{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.170537010159651,
  "eval_steps": 3000,
  "global_step": 57000,
  "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
    },
    {
      "entropy": 5.83179087638855,
      "epoch": 0.5447206095791002,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004976737668960034,
      "loss": 5.8127,
      "mean_token_accuracy": 0.14738722145557404,
      "num_tokens": 12483115.0,
      "step": 6005
    },
    {
      "entropy": 5.831946611404419,
      "epoch": 0.5451741654571843,
      "grad_norm": 2.0,
      "learning_rate": 0.0004976691238733688,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1467287912964821,
      "num_tokens": 12493280.0,
      "step": 6010
    },
    {
      "entropy": 5.849226713180542,
      "epoch": 0.5456277213352685,
      "grad_norm": 2.65625,
      "learning_rate": 0.00049766447624588,
      "loss": 5.7941,
      "mean_token_accuracy": 0.15283684581518173,
      "num_tokens": 12503665.0,
      "step": 6015
    },
    {
      "entropy": 5.815784883499146,
      "epoch": 0.5460812772133526,
      "grad_norm": 4.0625,
      "learning_rate": 0.000497659824013633,
      "loss": 5.7144,
      "mean_token_accuracy": 0.14608054608106613,
      "num_tokens": 12513381.0,
      "step": 6020
    },
    {
      "entropy": 5.895172071456909,
      "epoch": 0.5465348330914369,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004976551671767241,
      "loss": 5.754,
      "mean_token_accuracy": 0.1483724221587181,
      "num_tokens": 12524121.0,
      "step": 6025
    },
    {
      "entropy": 5.840110015869141,
      "epoch": 0.5469883889695211,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004976505057352496,
      "loss": 5.7072,
      "mean_token_accuracy": 0.15462427884340285,
      "num_tokens": 12533972.0,
      "step": 6030
    },
    {
      "entropy": 5.871407890319825,
      "epoch": 0.5474419448476052,
      "grad_norm": 5.8125,
      "learning_rate": 0.0004976458396893059,
      "loss": 5.8162,
      "mean_token_accuracy": 0.14339689910411835,
      "num_tokens": 12544110.0,
      "step": 6035
    },
    {
      "entropy": 5.787636804580688,
      "epoch": 0.5478955007256894,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004976411690389896,
      "loss": 5.6686,
      "mean_token_accuracy": 0.14848399460315703,
      "num_tokens": 12554273.0,
      "step": 6040
    },
    {
      "entropy": 5.903516960144043,
      "epoch": 0.5483490566037735,
      "grad_norm": 3.125,
      "learning_rate": 0.000497636493784397,
      "loss": 5.7501,
      "mean_token_accuracy": 0.1464612677693367,
      "num_tokens": 12564633.0,
      "step": 6045
    },
    {
      "entropy": 5.910893726348877,
      "epoch": 0.5488026124818578,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004976318139256251,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1580335780978203,
      "num_tokens": 12573790.0,
      "step": 6050
    },
    {
      "entropy": 5.878701543807983,
      "epoch": 0.549256168359942,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004976271294627706,
      "loss": 5.747,
      "mean_token_accuracy": 0.14373386204242705,
      "num_tokens": 12584265.0,
      "step": 6055
    },
    {
      "entropy": 5.770555639266968,
      "epoch": 0.5497097242380261,
      "grad_norm": 7.21875,
      "learning_rate": 0.0004976224403959303,
      "loss": 5.6169,
      "mean_token_accuracy": 0.15551466792821883,
      "num_tokens": 12594413.0,
      "step": 6060
    },
    {
      "entropy": 5.840236854553223,
      "epoch": 0.5501632801161103,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004976177467252012,
      "loss": 5.6874,
      "mean_token_accuracy": 0.15790211111307145,
      "num_tokens": 12604637.0,
      "step": 6065
    },
    {
      "entropy": 5.8618628025054935,
      "epoch": 0.5506168359941945,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004976130484506804,
      "loss": 5.7417,
      "mean_token_accuracy": 0.15392341166734697,
      "num_tokens": 12614873.0,
      "step": 6070
    },
    {
      "entropy": 5.796398115158081,
      "epoch": 0.5510703918722787,
      "grad_norm": 2.25,
      "learning_rate": 0.0004976083455724649,
      "loss": 5.6818,
      "mean_token_accuracy": 0.15129643827676773,
      "num_tokens": 12626182.0,
      "step": 6075
    },
    {
      "entropy": 5.792143869400024,
      "epoch": 0.5515239477503628,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004976036380906523,
      "loss": 5.6363,
      "mean_token_accuracy": 0.15263219848275184,
      "num_tokens": 12635418.0,
      "step": 6080
    },
    {
      "entropy": 5.859647369384765,
      "epoch": 0.551977503628447,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004975989260053395,
      "loss": 5.7718,
      "mean_token_accuracy": 0.1413577154278755,
      "num_tokens": 12645942.0,
      "step": 6085
    },
    {
      "entropy": 5.834799003601074,
      "epoch": 0.5524310595065312,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004975942093166242,
      "loss": 5.6965,
      "mean_token_accuracy": 0.1482281967997551,
      "num_tokens": 12656460.0,
      "step": 6090
    },
    {
      "entropy": 5.838398408889771,
      "epoch": 0.5528846153846154,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004975894880246041,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15041477978229523,
      "num_tokens": 12666916.0,
      "step": 6095
    },
    {
      "entropy": 5.82815203666687,
      "epoch": 0.5533381712626996,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004975847621293766,
      "loss": 5.7116,
      "mean_token_accuracy": 0.1567052870988846,
      "num_tokens": 12677562.0,
      "step": 6100
    },
    {
      "entropy": 5.815052461624146,
      "epoch": 0.5537917271407837,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004975800316310394,
      "loss": 5.6512,
      "mean_token_accuracy": 0.15303489565849304,
      "num_tokens": 12687349.0,
      "step": 6105
    },
    {
      "entropy": 5.88008337020874,
      "epoch": 0.5542452830188679,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004975752965296906,
      "loss": 5.7239,
      "mean_token_accuracy": 0.15200326591730118,
      "num_tokens": 12697850.0,
      "step": 6110
    },
    {
      "entropy": 5.805438089370727,
      "epoch": 0.5546988388969522,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004975705568254278,
      "loss": 5.5851,
      "mean_token_accuracy": 0.15197097063064574,
      "num_tokens": 12708283.0,
      "step": 6115
    },
    {
      "entropy": 5.785101699829101,
      "epoch": 0.5551523947750363,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004975658125183493,
      "loss": 5.6675,
      "mean_token_accuracy": 0.15603910833597184,
      "num_tokens": 12718201.0,
      "step": 6120
    },
    {
      "entropy": 5.77977442741394,
      "epoch": 0.5556059506531205,
      "grad_norm": 3.28125,
      "learning_rate": 0.000497561063608553,
      "loss": 5.6441,
      "mean_token_accuracy": 0.1591254323720932,
      "num_tokens": 12728860.0,
      "step": 6125
    },
    {
      "entropy": 5.882054996490479,
      "epoch": 0.5560595065312046,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004975563100961372,
      "loss": 5.7045,
      "mean_token_accuracy": 0.14953402429819107,
      "num_tokens": 12739007.0,
      "step": 6130
    },
    {
      "entropy": 5.851760959625244,
      "epoch": 0.5565130624092888,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004975515519812002,
      "loss": 5.5603,
      "mean_token_accuracy": 0.1571223884820938,
      "num_tokens": 12749004.0,
      "step": 6135
    },
    {
      "entropy": 5.719464206695557,
      "epoch": 0.556966618287373,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004975467892638402,
      "loss": 5.7312,
      "mean_token_accuracy": 0.15219081193208694,
      "num_tokens": 12760085.0,
      "step": 6140
    },
    {
      "entropy": 5.780846977233887,
      "epoch": 0.5574201741654572,
      "grad_norm": 2.125,
      "learning_rate": 0.0004975420219441561,
      "loss": 5.6557,
      "mean_token_accuracy": 0.15899507701396942,
      "num_tokens": 12770303.0,
      "step": 6145
    },
    {
      "entropy": 5.880466079711914,
      "epoch": 0.5578737300435413,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004975372500222462,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15486729443073272,
      "num_tokens": 12780334.0,
      "step": 6150
    },
    {
      "entropy": 5.8313295364379885,
      "epoch": 0.5583272859216255,
      "grad_norm": 2.5,
      "learning_rate": 0.0004975324734982094,
      "loss": 5.7253,
      "mean_token_accuracy": 0.1489449195563793,
      "num_tokens": 12790980.0,
      "step": 6155
    },
    {
      "entropy": 5.845129108428955,
      "epoch": 0.5587808417997098,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004975276923721442,
      "loss": 5.7754,
      "mean_token_accuracy": 0.14961862117052077,
      "num_tokens": 12801713.0,
      "step": 6160
    },
    {
      "entropy": 5.870355415344238,
      "epoch": 0.5592343976777939,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004975229066441496,
      "loss": 5.758,
      "mean_token_accuracy": 0.14657938480377197,
      "num_tokens": 12813040.0,
      "step": 6165
    },
    {
      "entropy": 5.833003377914428,
      "epoch": 0.5596879535558781,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004975181163143247,
      "loss": 5.713,
      "mean_token_accuracy": 0.1453935071825981,
      "num_tokens": 12824048.0,
      "step": 6170
    },
    {
      "entropy": 5.837733316421509,
      "epoch": 0.5601415094339622,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004975133213827684,
      "loss": 5.6621,
      "mean_token_accuracy": 0.14674486815929413,
      "num_tokens": 12834297.0,
      "step": 6175
    },
    {
      "entropy": 5.896747970581055,
      "epoch": 0.5605950653120464,
      "grad_norm": 1.75,
      "learning_rate": 0.0004975085218495798,
      "loss": 5.7316,
      "mean_token_accuracy": 0.1485936962068081,
      "num_tokens": 12844574.0,
      "step": 6180
    },
    {
      "entropy": 5.765360546112061,
      "epoch": 0.5610486211901307,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004975037177148585,
      "loss": 5.6848,
      "mean_token_accuracy": 0.15071796774864196,
      "num_tokens": 12854670.0,
      "step": 6185
    },
    {
      "entropy": 5.713732099533081,
      "epoch": 0.5615021770682148,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004974989089787035,
      "loss": 5.7079,
      "mean_token_accuracy": 0.15234477072954178,
      "num_tokens": 12865281.0,
      "step": 6190
    },
    {
      "entropy": 5.8670814514160154,
      "epoch": 0.561955732946299,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004974940956412143,
      "loss": 5.6978,
      "mean_token_accuracy": 0.1518659010529518,
      "num_tokens": 12875222.0,
      "step": 6195
    },
    {
      "entropy": 5.996776294708252,
      "epoch": 0.5624092888243831,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004974892777024905,
      "loss": 5.7535,
      "mean_token_accuracy": 0.14664066731929778,
      "num_tokens": 12885079.0,
      "step": 6200
    },
    {
      "entropy": 5.816704797744751,
      "epoch": 0.5628628447024674,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004974844551626318,
      "loss": 5.7058,
      "mean_token_accuracy": 0.15577784329652786,
      "num_tokens": 12896402.0,
      "step": 6205
    },
    {
      "entropy": 5.7654026508331295,
      "epoch": 0.5633164005805515,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004974796280217379,
      "loss": 5.6738,
      "mean_token_accuracy": 0.16155672520399095,
      "num_tokens": 12906403.0,
      "step": 6210
    },
    {
      "entropy": 5.881820344924927,
      "epoch": 0.5637699564586357,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004974747962799086,
      "loss": 5.694,
      "mean_token_accuracy": 0.14602868854999543,
      "num_tokens": 12917061.0,
      "step": 6215
    },
    {
      "entropy": 5.8594695091247555,
      "epoch": 0.5642235123367199,
      "grad_norm": 7.03125,
      "learning_rate": 0.0004974699599372437,
      "loss": 5.7495,
      "mean_token_accuracy": 0.15344917625188828,
      "num_tokens": 12927182.0,
      "step": 6220
    },
    {
      "entropy": 5.867999505996704,
      "epoch": 0.564677068214804,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004974651189938434,
      "loss": 5.683,
      "mean_token_accuracy": 0.15102926790714263,
      "num_tokens": 12937371.0,
      "step": 6225
    },
    {
      "entropy": 5.83708872795105,
      "epoch": 0.5651306240928883,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004974602734498079,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1440092071890831,
      "num_tokens": 12948569.0,
      "step": 6230
    },
    {
      "entropy": 5.8283039093017575,
      "epoch": 0.5655841799709724,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000497455423305237,
      "loss": 5.6769,
      "mean_token_accuracy": 0.14810833483934402,
      "num_tokens": 12958503.0,
      "step": 6235
    },
    {
      "entropy": 5.854265213012695,
      "epoch": 0.5660377358490566,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004974505685602314,
      "loss": 5.6917,
      "mean_token_accuracy": 0.15404328405857087,
      "num_tokens": 12968714.0,
      "step": 6240
    },
    {
      "entropy": 5.803040409088135,
      "epoch": 0.5664912917271407,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004974457092148912,
      "loss": 5.7329,
      "mean_token_accuracy": 0.14936806112527848,
      "num_tokens": 12978857.0,
      "step": 6245
    },
    {
      "entropy": 5.967836856842041,
      "epoch": 0.566944847605225,
      "grad_norm": 2.375,
      "learning_rate": 0.0004974408452693172,
      "loss": 5.8515,
      "mean_token_accuracy": 0.14415958821773528,
      "num_tokens": 12989863.0,
      "step": 6250
    },
    {
      "entropy": 5.897352266311645,
      "epoch": 0.5673984034833092,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004974359767236099,
      "loss": 5.648,
      "mean_token_accuracy": 0.15427448004484176,
      "num_tokens": 13000307.0,
      "step": 6255
    },
    {
      "entropy": 5.835937643051148,
      "epoch": 0.5678519593613933,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004974311035778698,
      "loss": 5.7076,
      "mean_token_accuracy": 0.16386753916740418,
      "num_tokens": 13009053.0,
      "step": 6260
    },
    {
      "entropy": 5.788226127624512,
      "epoch": 0.5683055152394775,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004974262258321979,
      "loss": 5.7386,
      "mean_token_accuracy": 0.14352438002824783,
      "num_tokens": 13019136.0,
      "step": 6265
    },
    {
      "entropy": 5.879953765869141,
      "epoch": 0.5687590711175616,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004974213434866949,
      "loss": 5.7667,
      "mean_token_accuracy": 0.15057471692562102,
      "num_tokens": 13030356.0,
      "step": 6270
    },
    {
      "entropy": 5.882538843154907,
      "epoch": 0.5692126269956459,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004974164565414617,
      "loss": 5.7889,
      "mean_token_accuracy": 0.1513417199254036,
      "num_tokens": 13040477.0,
      "step": 6275
    },
    {
      "entropy": 5.837933874130249,
      "epoch": 0.56966618287373,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004974115649965998,
      "loss": 5.6785,
      "mean_token_accuracy": 0.1549770399928093,
      "num_tokens": 13051755.0,
      "step": 6280
    },
    {
      "entropy": 5.841934442520142,
      "epoch": 0.5701197387518142,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004974066688522098,
      "loss": 5.6869,
      "mean_token_accuracy": 0.15811796933412553,
      "num_tokens": 13061670.0,
      "step": 6285
    },
    {
      "entropy": 5.760716295242309,
      "epoch": 0.5705732946298984,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004974017681083934,
      "loss": 5.6512,
      "mean_token_accuracy": 0.15475476533174515,
      "num_tokens": 13070917.0,
      "step": 6290
    },
    {
      "entropy": 5.800657749176025,
      "epoch": 0.5710268505079826,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004973968627652516,
      "loss": 5.7069,
      "mean_token_accuracy": 0.15006187707185745,
      "num_tokens": 13080577.0,
      "step": 6295
    },
    {
      "entropy": 5.851555442810058,
      "epoch": 0.5714804063860668,
      "grad_norm": 2.359375,
      "learning_rate": 0.000497391952822886,
      "loss": 5.7457,
      "mean_token_accuracy": 0.15279553458094597,
      "num_tokens": 13091446.0,
      "step": 6300
    },
    {
      "entropy": 5.947834491729736,
      "epoch": 0.5719339622641509,
      "grad_norm": 2.375,
      "learning_rate": 0.0004973870382813983,
      "loss": 5.7611,
      "mean_token_accuracy": 0.15199481695890427,
      "num_tokens": 13102752.0,
      "step": 6305
    },
    {
      "entropy": 5.940124607086181,
      "epoch": 0.5723875181422351,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004973821191408899,
      "loss": 5.7886,
      "mean_token_accuracy": 0.1465797521173954,
      "num_tokens": 13113944.0,
      "step": 6310
    },
    {
      "entropy": 5.7535113334655765,
      "epoch": 0.5728410740203193,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004973771954014626,
      "loss": 5.7191,
      "mean_token_accuracy": 0.14859745651483536,
      "num_tokens": 13124316.0,
      "step": 6315
    },
    {
      "entropy": 5.874121236801147,
      "epoch": 0.5732946298984035,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004973722670632182,
      "loss": 5.8198,
      "mean_token_accuracy": 0.14575577676296234,
      "num_tokens": 13133567.0,
      "step": 6320
    },
    {
      "entropy": 5.899265956878662,
      "epoch": 0.5737481857764877,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004973673341262587,
      "loss": 5.7334,
      "mean_token_accuracy": 0.14762221351265908,
      "num_tokens": 13144279.0,
      "step": 6325
    },
    {
      "entropy": 5.887506055831909,
      "epoch": 0.5742017416545718,
      "grad_norm": 2.671875,
      "learning_rate": 0.000497362396590686,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1508809007704258,
      "num_tokens": 13154463.0,
      "step": 6330
    },
    {
      "entropy": 5.8973757266998295,
      "epoch": 0.574655297532656,
      "grad_norm": 2.5,
      "learning_rate": 0.0004973574544566024,
      "loss": 5.7407,
      "mean_token_accuracy": 0.14477827101945878,
      "num_tokens": 13165565.0,
      "step": 6335
    },
    {
      "entropy": 5.804881763458252,
      "epoch": 0.5751088534107403,
      "grad_norm": 4.0,
      "learning_rate": 0.0004973525077241098,
      "loss": 5.6568,
      "mean_token_accuracy": 0.15822704434394835,
      "num_tokens": 13175839.0,
      "step": 6340
    },
    {
      "entropy": 5.780976009368897,
      "epoch": 0.5755624092888244,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004973475563933109,
      "loss": 5.6726,
      "mean_token_accuracy": 0.14872757494449615,
      "num_tokens": 13186372.0,
      "step": 6345
    },
    {
      "entropy": 5.8148088455200195,
      "epoch": 0.5760159651669086,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004973426004643077,
      "loss": 5.702,
      "mean_token_accuracy": 0.15282864421606063,
      "num_tokens": 13196259.0,
      "step": 6350
    },
    {
      "entropy": 5.822752189636231,
      "epoch": 0.5764695210449927,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004973376399372029,
      "loss": 5.694,
      "mean_token_accuracy": 0.15161322802305222,
      "num_tokens": 13206497.0,
      "step": 6355
    },
    {
      "entropy": 5.866645383834839,
      "epoch": 0.5769230769230769,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004973326748120991,
      "loss": 5.7283,
      "mean_token_accuracy": 0.15126238614320756,
      "num_tokens": 13217188.0,
      "step": 6360
    },
    {
      "entropy": 5.846437931060791,
      "epoch": 0.5773766328011611,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004973277050890988,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1515586942434311,
      "num_tokens": 13226544.0,
      "step": 6365
    },
    {
      "entropy": 5.839645147323608,
      "epoch": 0.5778301886792453,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004973227307683049,
      "loss": 5.7764,
      "mean_token_accuracy": 0.14674135893583298,
      "num_tokens": 13236572.0,
      "step": 6370
    },
    {
      "entropy": 5.9877115249633786,
      "epoch": 0.5782837445573294,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004973177518498204,
      "loss": 5.7779,
      "mean_token_accuracy": 0.14902501702308654,
      "num_tokens": 13247958.0,
      "step": 6375
    },
    {
      "entropy": 5.8630200862884525,
      "epoch": 0.5787373004354136,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004973127683337482,
      "loss": 5.6749,
      "mean_token_accuracy": 0.14949809312820433,
      "num_tokens": 13258255.0,
      "step": 6380
    },
    {
      "entropy": 5.744321727752686,
      "epoch": 0.5791908563134979,
      "grad_norm": 2.25,
      "learning_rate": 0.0004973077802201911,
      "loss": 5.6909,
      "mean_token_accuracy": 0.15299025177955627,
      "num_tokens": 13268416.0,
      "step": 6385
    },
    {
      "entropy": 5.801497316360473,
      "epoch": 0.579644412191582,
      "grad_norm": 2.875,
      "learning_rate": 0.0004973027875092526,
      "loss": 5.6943,
      "mean_token_accuracy": 0.15058967620134353,
      "num_tokens": 13278087.0,
      "step": 6390
    },
    {
      "entropy": 5.81386513710022,
      "epoch": 0.5800979680696662,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004972977902010357,
      "loss": 5.592,
      "mean_token_accuracy": 0.1598864108324051,
      "num_tokens": 13288121.0,
      "step": 6395
    },
    {
      "entropy": 5.926956272125244,
      "epoch": 0.5805515239477503,
      "grad_norm": 2.25,
      "learning_rate": 0.0004972927882956439,
      "loss": 5.7942,
      "mean_token_accuracy": 0.14235642403364182,
      "num_tokens": 13298620.0,
      "step": 6400
    },
    {
      "entropy": 5.7990469455719,
      "epoch": 0.5810050798258345,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004972877817931807,
      "loss": 5.6142,
      "mean_token_accuracy": 0.14995205849409105,
      "num_tokens": 13308413.0,
      "step": 6405
    },
    {
      "entropy": 5.788681268692017,
      "epoch": 0.5814586357039188,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004972827706937493,
      "loss": 5.7003,
      "mean_token_accuracy": 0.1515623390674591,
      "num_tokens": 13319387.0,
      "step": 6410
    },
    {
      "entropy": 5.825255298614502,
      "epoch": 0.5819121915820029,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004972777549974537,
      "loss": 5.6635,
      "mean_token_accuracy": 0.1523427128791809,
      "num_tokens": 13330494.0,
      "step": 6415
    },
    {
      "entropy": 5.863323593139649,
      "epoch": 0.5823657474600871,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004972727347043975,
      "loss": 5.663,
      "mean_token_accuracy": 0.15184202492237092,
      "num_tokens": 13341009.0,
      "step": 6420
    },
    {
      "entropy": 5.795520925521851,
      "epoch": 0.5828193033381712,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004972677098146845,
      "loss": 5.6168,
      "mean_token_accuracy": 0.15109016746282578,
      "num_tokens": 13351269.0,
      "step": 6425
    },
    {
      "entropy": 5.784488582611084,
      "epoch": 0.5832728592162555,
      "grad_norm": 2.625,
      "learning_rate": 0.0004972626803284185,
      "loss": 5.7503,
      "mean_token_accuracy": 0.14944482147693633,
      "num_tokens": 13362160.0,
      "step": 6430
    },
    {
      "entropy": 5.889904022216797,
      "epoch": 0.5837264150943396,
      "grad_norm": 3.375,
      "learning_rate": 0.0004972576462457038,
      "loss": 5.7866,
      "mean_token_accuracy": 0.14930133521556854,
      "num_tokens": 13372497.0,
      "step": 6435
    },
    {
      "entropy": 5.957094621658325,
      "epoch": 0.5841799709724238,
      "grad_norm": 2.125,
      "learning_rate": 0.0004972526075666442,
      "loss": 5.7257,
      "mean_token_accuracy": 0.15872014164924622,
      "num_tokens": 13381665.0,
      "step": 6440
    },
    {
      "entropy": 5.8321576595306395,
      "epoch": 0.584633526850508,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004972475642913442,
      "loss": 5.6392,
      "mean_token_accuracy": 0.1554955556988716,
      "num_tokens": 13392507.0,
      "step": 6445
    },
    {
      "entropy": 5.76501989364624,
      "epoch": 0.5850870827285921,
      "grad_norm": 3.5,
      "learning_rate": 0.0004972425164199078,
      "loss": 5.6346,
      "mean_token_accuracy": 0.1577868416905403,
      "num_tokens": 13402171.0,
      "step": 6450
    },
    {
      "entropy": 5.8788012027740475,
      "epoch": 0.5855406386066764,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004972374639524396,
      "loss": 5.8384,
      "mean_token_accuracy": 0.14709555059671403,
      "num_tokens": 13412933.0,
      "step": 6455
    },
    {
      "entropy": 5.856866788864136,
      "epoch": 0.5859941944847605,
      "grad_norm": 2.15625,
      "learning_rate": 0.000497232406889044,
      "loss": 5.6662,
      "mean_token_accuracy": 0.1488827720284462,
      "num_tokens": 13423563.0,
      "step": 6460
    },
    {
      "entropy": 5.896422910690307,
      "epoch": 0.5864477503628447,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004972273452298255,
      "loss": 5.6954,
      "mean_token_accuracy": 0.15475627183914184,
      "num_tokens": 13433998.0,
      "step": 6465
    },
    {
      "entropy": 5.828073978424072,
      "epoch": 0.5869013062409288,
      "grad_norm": 2.375,
      "learning_rate": 0.0004972222789748888,
      "loss": 5.6352,
      "mean_token_accuracy": 0.15622286945581437,
      "num_tokens": 13443878.0,
      "step": 6470
    },
    {
      "entropy": 5.796691513061523,
      "epoch": 0.5873548621190131,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004972172081243389,
      "loss": 5.7249,
      "mean_token_accuracy": 0.1440662868320942,
      "num_tokens": 13456051.0,
      "step": 6475
    },
    {
      "entropy": 5.815340518951416,
      "epoch": 0.5878084179970973,
      "grad_norm": 4.375,
      "learning_rate": 0.0004972121326782806,
      "loss": 5.7373,
      "mean_token_accuracy": 0.15519822090864183,
      "num_tokens": 13467276.0,
      "step": 6480
    },
    {
      "entropy": 5.861797666549682,
      "epoch": 0.5882619738751814,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004972070526368186,
      "loss": 5.7039,
      "mean_token_accuracy": 0.14981682896614074,
      "num_tokens": 13478331.0,
      "step": 6485
    },
    {
      "entropy": 5.927264070510864,
      "epoch": 0.5887155297532656,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004972019680000582,
      "loss": 5.6366,
      "mean_token_accuracy": 0.1500559240579605,
      "num_tokens": 13489879.0,
      "step": 6490
    },
    {
      "entropy": 5.851372814178466,
      "epoch": 0.5891690856313497,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004971968787681044,
      "loss": 5.763,
      "mean_token_accuracy": 0.1548596978187561,
      "num_tokens": 13500879.0,
      "step": 6495
    },
    {
      "entropy": 5.862674856185913,
      "epoch": 0.589622641509434,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004971917849410627,
      "loss": 5.7439,
      "mean_token_accuracy": 0.14779864698648454,
      "num_tokens": 13511719.0,
      "step": 6500
    },
    {
      "entropy": 5.834804105758667,
      "epoch": 0.5900761973875182,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004971866865190381,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1496259018778801,
      "num_tokens": 13521850.0,
      "step": 6505
    },
    {
      "entropy": 5.923253965377808,
      "epoch": 0.5905297532656023,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004971815835021362,
      "loss": 5.6981,
      "mean_token_accuracy": 0.15618499219417573,
      "num_tokens": 13532095.0,
      "step": 6510
    },
    {
      "entropy": 5.815547132492066,
      "epoch": 0.5909833091436865,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004971764758904626,
      "loss": 5.6828,
      "mean_token_accuracy": 0.14781468808650972,
      "num_tokens": 13543133.0,
      "step": 6515
    },
    {
      "entropy": 5.811095142364502,
      "epoch": 0.5914368650217707,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004971713636841229,
      "loss": 5.7401,
      "mean_token_accuracy": 0.14562103599309922,
      "num_tokens": 13553771.0,
      "step": 6520
    },
    {
      "entropy": 5.810519790649414,
      "epoch": 0.5918904208998549,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004971662468832228,
      "loss": 5.6905,
      "mean_token_accuracy": 0.15237817242741586,
      "num_tokens": 13564548.0,
      "step": 6525
    },
    {
      "entropy": 5.833587503433227,
      "epoch": 0.592343976777939,
      "grad_norm": 2.265625,
      "learning_rate": 0.000497161125487868,
      "loss": 5.6855,
      "mean_token_accuracy": 0.15271230041980743,
      "num_tokens": 13574567.0,
      "step": 6530
    },
    {
      "entropy": 5.78583288192749,
      "epoch": 0.5927975326560232,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004971559994981645,
      "loss": 5.7089,
      "mean_token_accuracy": 0.1522444397211075,
      "num_tokens": 13585302.0,
      "step": 6535
    },
    {
      "entropy": 5.878629112243653,
      "epoch": 0.5932510885341074,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004971508689142184,
      "loss": 5.7447,
      "mean_token_accuracy": 0.14830908924341202,
      "num_tokens": 13594754.0,
      "step": 6540
    },
    {
      "entropy": 5.83653244972229,
      "epoch": 0.5937046444121916,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004971457337361356,
      "loss": 5.6665,
      "mean_token_accuracy": 0.156255804002285,
      "num_tokens": 13604691.0,
      "step": 6545
    },
    {
      "entropy": 5.710748481750488,
      "epoch": 0.5941582002902758,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004971405939640225,
      "loss": 5.5835,
      "mean_token_accuracy": 0.15645805299282073,
      "num_tokens": 13613911.0,
      "step": 6550
    },
    {
      "entropy": 5.8865584373474125,
      "epoch": 0.5946117561683599,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004971354495979854,
      "loss": 5.7821,
      "mean_token_accuracy": 0.15241619274020196,
      "num_tokens": 13624285.0,
      "step": 6555
    },
    {
      "entropy": 5.771980142593383,
      "epoch": 0.5950653120464441,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004971303006381305,
      "loss": 5.6256,
      "mean_token_accuracy": 0.1564767062664032,
      "num_tokens": 13635434.0,
      "step": 6560
    },
    {
      "entropy": 5.796686601638794,
      "epoch": 0.5955188679245284,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004971251470845643,
      "loss": 5.6973,
      "mean_token_accuracy": 0.15278116166591643,
      "num_tokens": 13645582.0,
      "step": 6565
    },
    {
      "entropy": 5.8198816776275635,
      "epoch": 0.5959724238026125,
      "grad_norm": 2.875,
      "learning_rate": 0.0004971199889373935,
      "loss": 5.6479,
      "mean_token_accuracy": 0.15285747945308686,
      "num_tokens": 13655029.0,
      "step": 6570
    },
    {
      "entropy": 5.809751796722412,
      "epoch": 0.5964259796806967,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004971148261967249,
      "loss": 5.6286,
      "mean_token_accuracy": 0.15282736122608184,
      "num_tokens": 13665204.0,
      "step": 6575
    },
    {
      "entropy": 5.816667556762695,
      "epoch": 0.5968795355587808,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004971096588626649,
      "loss": 5.7408,
      "mean_token_accuracy": 0.15095006227493285,
      "num_tokens": 13676281.0,
      "step": 6580
    },
    {
      "entropy": 5.780574703216553,
      "epoch": 0.597333091436865,
      "grad_norm": 2.25,
      "learning_rate": 0.0004971044869353206,
      "loss": 5.7281,
      "mean_token_accuracy": 0.15921155661344527,
      "num_tokens": 13687144.0,
      "step": 6585
    },
    {
      "entropy": 5.795083045959473,
      "epoch": 0.5977866473149492,
      "grad_norm": 1.953125,
      "learning_rate": 0.000497099310414799,
      "loss": 5.688,
      "mean_token_accuracy": 0.15438786298036575,
      "num_tokens": 13696371.0,
      "step": 6590
    },
    {
      "entropy": 5.84175672531128,
      "epoch": 0.5982402031930334,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004970941293012069,
      "loss": 5.6345,
      "mean_token_accuracy": 0.15077201277017593,
      "num_tokens": 13706900.0,
      "step": 6595
    },
    {
      "entropy": 5.836261129379272,
      "epoch": 0.5986937590711175,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004970889435946518,
      "loss": 5.7144,
      "mean_token_accuracy": 0.15443189740180968,
      "num_tokens": 13717874.0,
      "step": 6600
    },
    {
      "entropy": 5.776962184906006,
      "epoch": 0.5991473149492017,
      "grad_norm": 2.75,
      "learning_rate": 0.0004970837532952406,
      "loss": 5.6737,
      "mean_token_accuracy": 0.1497507020831108,
      "num_tokens": 13728020.0,
      "step": 6605
    },
    {
      "entropy": 5.830104970932007,
      "epoch": 0.599600870827286,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004970785584030808,
      "loss": 5.7138,
      "mean_token_accuracy": 0.1525508925318718,
      "num_tokens": 13738007.0,
      "step": 6610
    },
    {
      "entropy": 5.879718828201294,
      "epoch": 0.6000544267053701,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004970733589182799,
      "loss": 5.736,
      "mean_token_accuracy": 0.1473026007413864,
      "num_tokens": 13749429.0,
      "step": 6615
    },
    {
      "entropy": 5.8097755908966064,
      "epoch": 0.6005079825834543,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004970681548409452,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1569398522377014,
      "num_tokens": 13760279.0,
      "step": 6620
    },
    {
      "entropy": 5.83681287765503,
      "epoch": 0.6009615384615384,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004970629461711846,
      "loss": 5.8122,
      "mean_token_accuracy": 0.14485863000154495,
      "num_tokens": 13770575.0,
      "step": 6625
    },
    {
      "entropy": 5.8300028324127195,
      "epoch": 0.6014150943396226,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004970577329091057,
      "loss": 5.6667,
      "mean_token_accuracy": 0.15231173038482665,
      "num_tokens": 13781068.0,
      "step": 6630
    },
    {
      "entropy": 5.8861555576324465,
      "epoch": 0.6018686502177069,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004970525150548162,
      "loss": 5.7296,
      "mean_token_accuracy": 0.15105716288089752,
      "num_tokens": 13790745.0,
      "step": 6635
    },
    {
      "entropy": 5.805358266830444,
      "epoch": 0.602322206095791,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004970472926084242,
      "loss": 5.6751,
      "mean_token_accuracy": 0.1507732406258583,
      "num_tokens": 13800196.0,
      "step": 6640
    },
    {
      "entropy": 5.770063829421997,
      "epoch": 0.6027757619738752,
      "grad_norm": 16.875,
      "learning_rate": 0.0004970420655700375,
      "loss": 5.5856,
      "mean_token_accuracy": 0.15864112824201584,
      "num_tokens": 13809954.0,
      "step": 6645
    },
    {
      "entropy": 5.792227268218994,
      "epoch": 0.6032293178519593,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004970368339397643,
      "loss": 5.7318,
      "mean_token_accuracy": 0.14497790783643721,
      "num_tokens": 13820281.0,
      "step": 6650
    },
    {
      "entropy": 5.885860538482666,
      "epoch": 0.6036828737300436,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004970315977177129,
      "loss": 5.7127,
      "mean_token_accuracy": 0.15290192514657974,
      "num_tokens": 13830614.0,
      "step": 6655
    },
    {
      "entropy": 5.7723931789398195,
      "epoch": 0.6041364296081277,
      "grad_norm": 8.3125,
      "learning_rate": 0.0004970263569039914,
      "loss": 5.6539,
      "mean_token_accuracy": 0.15475403964519502,
      "num_tokens": 13840551.0,
      "step": 6660
    },
    {
      "entropy": 5.899419689178467,
      "epoch": 0.6045899854862119,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004970211114987083,
      "loss": 5.8363,
      "mean_token_accuracy": 0.14742269665002822,
      "num_tokens": 13851830.0,
      "step": 6665
    },
    {
      "entropy": 5.820240592956543,
      "epoch": 0.6050435413642961,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004970158615019719,
      "loss": 5.6443,
      "mean_token_accuracy": 0.15448153614997864,
      "num_tokens": 13862299.0,
      "step": 6670
    },
    {
      "entropy": 5.92971625328064,
      "epoch": 0.6054970972423802,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004970106069138912,
      "loss": 5.769,
      "mean_token_accuracy": 0.14795244932174684,
      "num_tokens": 13872852.0,
      "step": 6675
    },
    {
      "entropy": 5.852387952804565,
      "epoch": 0.6059506531204645,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004970053477345743,
      "loss": 5.7018,
      "mean_token_accuracy": 0.16219263672828674,
      "num_tokens": 13883601.0,
      "step": 6680
    },
    {
      "entropy": 5.804964447021485,
      "epoch": 0.6064042089985486,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004970000839641304,
      "loss": 5.709,
      "mean_token_accuracy": 0.15332725495100022,
      "num_tokens": 13894246.0,
      "step": 6685
    },
    {
      "entropy": 5.776219987869263,
      "epoch": 0.6068577648766328,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004969948156026681,
      "loss": 5.7196,
      "mean_token_accuracy": 0.15061549395322799,
      "num_tokens": 13904823.0,
      "step": 6690
    },
    {
      "entropy": 5.924747323989868,
      "epoch": 0.6073113207547169,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004969895426502965,
      "loss": 5.7394,
      "mean_token_accuracy": 0.147009214758873,
      "num_tokens": 13915338.0,
      "step": 6695
    },
    {
      "entropy": 5.8624742984771725,
      "epoch": 0.6077648766328012,
      "grad_norm": 9.75,
      "learning_rate": 0.0004969842651071246,
      "loss": 5.6918,
      "mean_token_accuracy": 0.15186342597007751,
      "num_tokens": 13925838.0,
      "step": 6700
    },
    {
      "entropy": 5.8716295719146725,
      "epoch": 0.6082184325108854,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004969789829732615,
      "loss": 5.6917,
      "mean_token_accuracy": 0.14943995475769042,
      "num_tokens": 13936341.0,
      "step": 6705
    },
    {
      "entropy": 5.775670528411865,
      "epoch": 0.6086719883889695,
      "grad_norm": 2.5,
      "learning_rate": 0.0004969736962488164,
      "loss": 5.7052,
      "mean_token_accuracy": 0.15158903300762178,
      "num_tokens": 13945438.0,
      "step": 6710
    },
    {
      "entropy": 5.7914879322052,
      "epoch": 0.6091255442670537,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004969684049338987,
      "loss": 5.6138,
      "mean_token_accuracy": 0.1479756087064743,
      "num_tokens": 13956029.0,
      "step": 6715
    },
    {
      "entropy": 5.867767429351806,
      "epoch": 0.6095791001451378,
      "grad_norm": 2.21875,
      "learning_rate": 0.000496963109028618,
      "loss": 5.7231,
      "mean_token_accuracy": 0.14818116053938865,
      "num_tokens": 13966498.0,
      "step": 6720
    },
    {
      "entropy": 5.8269449234008786,
      "epoch": 0.6100326560232221,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004969578085330834,
      "loss": 5.6769,
      "mean_token_accuracy": 0.149531489610672,
      "num_tokens": 13976997.0,
      "step": 6725
    },
    {
      "entropy": 5.888552141189575,
      "epoch": 0.6104862119013063,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004969525034474049,
      "loss": 5.6642,
      "mean_token_accuracy": 0.15069333463907242,
      "num_tokens": 13986664.0,
      "step": 6730
    },
    {
      "entropy": 5.795527458190918,
      "epoch": 0.6109397677793904,
      "grad_norm": 2.984375,
      "learning_rate": 0.000496947193771692,
      "loss": 5.6811,
      "mean_token_accuracy": 0.1521475911140442,
      "num_tokens": 13997828.0,
      "step": 6735
    },
    {
      "entropy": 5.740749216079712,
      "epoch": 0.6113933236574746,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004969418795060548,
      "loss": 5.6303,
      "mean_token_accuracy": 0.15071731954813003,
      "num_tokens": 14008293.0,
      "step": 6740
    },
    {
      "entropy": 5.788277959823608,
      "epoch": 0.6118468795355588,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004969365606506027,
      "loss": 5.6121,
      "mean_token_accuracy": 0.15342239588499068,
      "num_tokens": 14018031.0,
      "step": 6745
    },
    {
      "entropy": 5.80244402885437,
      "epoch": 0.612300435413643,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004969312372054461,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1577866330742836,
      "num_tokens": 14028013.0,
      "step": 6750
    },
    {
      "entropy": 5.764116668701172,
      "epoch": 0.6127539912917271,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004969259091706948,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1519751712679863,
      "num_tokens": 14039787.0,
      "step": 6755
    },
    {
      "entropy": 5.7666487216949465,
      "epoch": 0.6132075471698113,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004969205765464593,
      "loss": 5.6029,
      "mean_token_accuracy": 0.1544177196919918,
      "num_tokens": 14049223.0,
      "step": 6760
    },
    {
      "entropy": 5.835097169876098,
      "epoch": 0.6136611030478955,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004969152393328497,
      "loss": 5.6536,
      "mean_token_accuracy": 0.15525499433279039,
      "num_tokens": 14059004.0,
      "step": 6765
    },
    {
      "entropy": 5.888670158386231,
      "epoch": 0.6141146589259797,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004969098975299764,
      "loss": 5.7625,
      "mean_token_accuracy": 0.14184020161628724,
      "num_tokens": 14069615.0,
      "step": 6770
    },
    {
      "entropy": 5.841464519500732,
      "epoch": 0.6145682148040639,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004969045511379499,
      "loss": 5.6905,
      "mean_token_accuracy": 0.15196973383426665,
      "num_tokens": 14078886.0,
      "step": 6775
    },
    {
      "entropy": 5.859313821792602,
      "epoch": 0.615021770682148,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004968992001568806,
      "loss": 5.7616,
      "mean_token_accuracy": 0.14708426594734192,
      "num_tokens": 14089021.0,
      "step": 6780
    },
    {
      "entropy": 5.780261850357055,
      "epoch": 0.6154753265602322,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004968938445868795,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15852924063801765,
      "num_tokens": 14098316.0,
      "step": 6785
    },
    {
      "entropy": 5.832832527160645,
      "epoch": 0.6159288824383164,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004968884844280569,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15131434947252273,
      "num_tokens": 14108874.0,
      "step": 6790
    },
    {
      "entropy": 5.833796405792237,
      "epoch": 0.6163824383164006,
      "grad_norm": 3.0,
      "learning_rate": 0.0004968831196805242,
      "loss": 5.7507,
      "mean_token_accuracy": 0.15231581628322602,
      "num_tokens": 14119613.0,
      "step": 6795
    },
    {
      "entropy": 5.878158283233643,
      "epoch": 0.6168359941944848,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004968777503443918,
      "loss": 5.6809,
      "mean_token_accuracy": 0.15027406364679335,
      "num_tokens": 14129085.0,
      "step": 6800
    },
    {
      "entropy": 5.774784278869629,
      "epoch": 0.6172895500725689,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004968723764197711,
      "loss": 5.5675,
      "mean_token_accuracy": 0.16327838152647017,
      "num_tokens": 14138461.0,
      "step": 6805
    },
    {
      "entropy": 5.768008708953857,
      "epoch": 0.6177431059506531,
      "grad_norm": 2.875,
      "learning_rate": 0.000496866997906773,
      "loss": 5.6372,
      "mean_token_accuracy": 0.161294561624527,
      "num_tokens": 14149126.0,
      "step": 6810
    },
    {
      "entropy": 5.774297380447388,
      "epoch": 0.6181966618287373,
      "grad_norm": 4.25,
      "learning_rate": 0.0004968616148055089,
      "loss": 5.6585,
      "mean_token_accuracy": 0.15676031559705733,
      "num_tokens": 14160068.0,
      "step": 6815
    },
    {
      "entropy": 5.860025787353516,
      "epoch": 0.6186502177068215,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004968562271160901,
      "loss": 5.7546,
      "mean_token_accuracy": 0.1506920039653778,
      "num_tokens": 14170288.0,
      "step": 6820
    },
    {
      "entropy": 5.866980457305909,
      "epoch": 0.6191037735849056,
      "grad_norm": 2.09375,
      "learning_rate": 0.000496850834838628,
      "loss": 5.7489,
      "mean_token_accuracy": 0.15387054532766342,
      "num_tokens": 14181231.0,
      "step": 6825
    },
    {
      "entropy": 5.823310661315918,
      "epoch": 0.6195573294629898,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004968454379732342,
      "loss": 5.6475,
      "mean_token_accuracy": 0.15675568729639053,
      "num_tokens": 14191715.0,
      "step": 6830
    },
    {
      "entropy": 5.823959398269653,
      "epoch": 0.6200108853410741,
      "grad_norm": 2.046875,
      "learning_rate": 0.00049684003652002,
      "loss": 5.6493,
      "mean_token_accuracy": 0.15095681548118592,
      "num_tokens": 14203416.0,
      "step": 6835
    },
    {
      "entropy": 5.77118558883667,
      "epoch": 0.6204644412191582,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004968346304790973,
      "loss": 5.6142,
      "mean_token_accuracy": 0.15649014487862586,
      "num_tokens": 14213658.0,
      "step": 6840
    },
    {
      "entropy": 5.8414740562438965,
      "epoch": 0.6209179970972424,
      "grad_norm": 2.5,
      "learning_rate": 0.0004968292198505781,
      "loss": 5.7781,
      "mean_token_accuracy": 0.1473475828766823,
      "num_tokens": 14223956.0,
      "step": 6845
    },
    {
      "entropy": 5.85752501487732,
      "epoch": 0.6213715529753265,
      "grad_norm": 3.140625,
      "learning_rate": 0.000496823804634574,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1460169270634651,
      "num_tokens": 14236801.0,
      "step": 6850
    },
    {
      "entropy": 5.755843448638916,
      "epoch": 0.6218251088534107,
      "grad_norm": 2.875,
      "learning_rate": 0.0004968183848311972,
      "loss": 5.5906,
      "mean_token_accuracy": 0.15509454011917115,
      "num_tokens": 14247382.0,
      "step": 6855
    },
    {
      "entropy": 5.873498058319091,
      "epoch": 0.622278664731495,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004968129604405596,
      "loss": 5.6613,
      "mean_token_accuracy": 0.15263956636190415,
      "num_tokens": 14258627.0,
      "step": 6860
    },
    {
      "entropy": 5.784880971908569,
      "epoch": 0.6227322206095791,
      "grad_norm": 2.5,
      "learning_rate": 0.0004968075314627734,
      "loss": 5.6584,
      "mean_token_accuracy": 0.15659964829683304,
      "num_tokens": 14269013.0,
      "step": 6865
    },
    {
      "entropy": 5.7966248989105225,
      "epoch": 0.6231857764876633,
      "grad_norm": 3.34375,
      "learning_rate": 0.000496802097897951,
      "loss": 5.6326,
      "mean_token_accuracy": 0.15666757971048356,
      "num_tokens": 14279620.0,
      "step": 6870
    },
    {
      "entropy": 5.860994577407837,
      "epoch": 0.6236393323657474,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004967966597462046,
      "loss": 5.6651,
      "mean_token_accuracy": 0.16124931573867798,
      "num_tokens": 14289448.0,
      "step": 6875
    },
    {
      "entropy": 5.885032081604004,
      "epoch": 0.6240928882438317,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004967912170076469,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15155680626630783,
      "num_tokens": 14299999.0,
      "step": 6880
    },
    {
      "entropy": 5.80174822807312,
      "epoch": 0.6245464441219158,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004967857696823901,
      "loss": 5.6443,
      "mean_token_accuracy": 0.14934637099504472,
      "num_tokens": 14310384.0,
      "step": 6885
    },
    {
      "entropy": 5.7569050788879395,
      "epoch": 0.625,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004967803177705472,
      "loss": 5.7437,
      "mean_token_accuracy": 0.14541705399751664,
      "num_tokens": 14320644.0,
      "step": 6890
    },
    {
      "entropy": 5.833973693847656,
      "epoch": 0.6254535558780842,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004967748612722308,
      "loss": 5.7003,
      "mean_token_accuracy": 0.15331256091594697,
      "num_tokens": 14330077.0,
      "step": 6895
    },
    {
      "entropy": 5.9259500980377195,
      "epoch": 0.6259071117561683,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004967694001875538,
      "loss": 5.8098,
      "mean_token_accuracy": 0.14541816860437393,
      "num_tokens": 14341549.0,
      "step": 6900
    },
    {
      "entropy": 5.868324947357178,
      "epoch": 0.6263606676342526,
      "grad_norm": 2.984375,
      "learning_rate": 0.000496763934516629,
      "loss": 5.6829,
      "mean_token_accuracy": 0.1533267766237259,
      "num_tokens": 14352146.0,
      "step": 6905
    },
    {
      "entropy": 5.799675416946411,
      "epoch": 0.6268142235123367,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004967584642595696,
      "loss": 5.6725,
      "mean_token_accuracy": 0.14867498576641083,
      "num_tokens": 14362521.0,
      "step": 6910
    },
    {
      "entropy": 5.820127725601196,
      "epoch": 0.6272677793904209,
      "grad_norm": 4.25,
      "learning_rate": 0.0004967529894164887,
      "loss": 5.7218,
      "mean_token_accuracy": 0.15302577912807463,
      "num_tokens": 14373216.0,
      "step": 6915
    },
    {
      "entropy": 5.8009356498718265,
      "epoch": 0.627721335268505,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004967475099874992,
      "loss": 5.664,
      "mean_token_accuracy": 0.1593104839324951,
      "num_tokens": 14383104.0,
      "step": 6920
    },
    {
      "entropy": 5.81319580078125,
      "epoch": 0.6281748911465893,
      "grad_norm": 3.1875,
      "learning_rate": 0.000496742025972715,
      "loss": 5.6722,
      "mean_token_accuracy": 0.16146908774971963,
      "num_tokens": 14393854.0,
      "step": 6925
    },
    {
      "entropy": 5.859877109527588,
      "epoch": 0.6286284470246735,
      "grad_norm": 2.109375,
      "learning_rate": 0.000496736537372249,
      "loss": 5.6132,
      "mean_token_accuracy": 0.16212746649980544,
      "num_tokens": 14403213.0,
      "step": 6930
    },
    {
      "entropy": 5.876523447036743,
      "epoch": 0.6290820029027576,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004967310441862151,
      "loss": 5.8071,
      "mean_token_accuracy": 0.14938885420560838,
      "num_tokens": 14412792.0,
      "step": 6935
    },
    {
      "entropy": 5.8900281429290775,
      "epoch": 0.6295355587808418,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004967255464147265,
      "loss": 5.6745,
      "mean_token_accuracy": 0.1568540945649147,
      "num_tokens": 14423061.0,
      "step": 6940
    },
    {
      "entropy": 5.874559831619263,
      "epoch": 0.6299891146589259,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004967200440578972,
      "loss": 5.6251,
      "mean_token_accuracy": 0.16184545755386354,
      "num_tokens": 14433338.0,
      "step": 6945
    },
    {
      "entropy": 5.755834102630615,
      "epoch": 0.6304426705370102,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004967145371158408,
      "loss": 5.6117,
      "mean_token_accuracy": 0.16235730051994324,
      "num_tokens": 14444124.0,
      "step": 6950
    },
    {
      "entropy": 5.839078283309936,
      "epoch": 0.6308962264150944,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004967090255886714,
      "loss": 5.7494,
      "mean_token_accuracy": 0.15068836957216264,
      "num_tokens": 14453868.0,
      "step": 6955
    },
    {
      "entropy": 5.887261533737183,
      "epoch": 0.6313497822931785,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004967035094765028,
      "loss": 5.6452,
      "mean_token_accuracy": 0.14719320237636566,
      "num_tokens": 14465015.0,
      "step": 6960
    },
    {
      "entropy": 5.804389810562133,
      "epoch": 0.6318033381712627,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004966979887794491,
      "loss": 5.6197,
      "mean_token_accuracy": 0.16230510920286179,
      "num_tokens": 14475523.0,
      "step": 6965
    },
    {
      "entropy": 5.753247451782227,
      "epoch": 0.6322568940493469,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004966924634976245,
      "loss": 5.6644,
      "mean_token_accuracy": 0.1508348599076271,
      "num_tokens": 14485523.0,
      "step": 6970
    },
    {
      "entropy": 5.880261182785034,
      "epoch": 0.6327104499274311,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004966869336311433,
      "loss": 5.7506,
      "mean_token_accuracy": 0.15484288036823274,
      "num_tokens": 14495237.0,
      "step": 6975
    },
    {
      "entropy": 5.907916021347046,
      "epoch": 0.6331640058055152,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004966813991801197,
      "loss": 5.8437,
      "mean_token_accuracy": 0.14812066629529,
      "num_tokens": 14506431.0,
      "step": 6980
    },
    {
      "entropy": 5.925472545623779,
      "epoch": 0.6336175616835994,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004966758601446684,
      "loss": 5.7761,
      "mean_token_accuracy": 0.1462002858519554,
      "num_tokens": 14517445.0,
      "step": 6985
    },
    {
      "entropy": 5.854861927032471,
      "epoch": 0.6340711175616836,
      "grad_norm": 4.75,
      "learning_rate": 0.0004966703165249038,
      "loss": 5.6887,
      "mean_token_accuracy": 0.14496298283338546,
      "num_tokens": 14527818.0,
      "step": 6990
    },
    {
      "entropy": 5.769768190383911,
      "epoch": 0.6345246734397678,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004966647683209406,
      "loss": 5.7801,
      "mean_token_accuracy": 0.14265136122703553,
      "num_tokens": 14538235.0,
      "step": 6995
    },
    {
      "entropy": 5.903887510299683,
      "epoch": 0.634978229317852,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004966592155328935,
      "loss": 5.7874,
      "mean_token_accuracy": 0.1466798961162567,
      "num_tokens": 14549784.0,
      "step": 7000
    },
    {
      "entropy": 5.886717748641968,
      "epoch": 0.6354317851959361,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004966536581608771,
      "loss": 5.7825,
      "mean_token_accuracy": 0.14561346769332886,
      "num_tokens": 14561057.0,
      "step": 7005
    },
    {
      "entropy": 5.904143190383911,
      "epoch": 0.6358853410740203,
      "grad_norm": 2.125,
      "learning_rate": 0.0004966480962050069,
      "loss": 5.7119,
      "mean_token_accuracy": 0.1525163695216179,
      "num_tokens": 14570859.0,
      "step": 7010
    },
    {
      "entropy": 5.81612000465393,
      "epoch": 0.6363388969521045,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004966425296653974,
      "loss": 5.6954,
      "mean_token_accuracy": 0.15854416489601136,
      "num_tokens": 14581066.0,
      "step": 7015
    },
    {
      "entropy": 5.780066299438476,
      "epoch": 0.6367924528301887,
      "grad_norm": 17.5,
      "learning_rate": 0.000496636958542164,
      "loss": 5.6691,
      "mean_token_accuracy": 0.15080804228782654,
      "num_tokens": 14590709.0,
      "step": 7020
    },
    {
      "entropy": 5.813197469711303,
      "epoch": 0.6372460087082729,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004966313828354218,
      "loss": 5.666,
      "mean_token_accuracy": 0.15459553748369217,
      "num_tokens": 14600849.0,
      "step": 7025
    },
    {
      "entropy": 5.7914598941802975,
      "epoch": 0.637699564586357,
      "grad_norm": 5.125,
      "learning_rate": 0.000496625802545286,
      "loss": 5.5547,
      "mean_token_accuracy": 0.15562158226966857,
      "num_tokens": 14611427.0,
      "step": 7030
    },
    {
      "entropy": 5.883208084106445,
      "epoch": 0.6381531204644412,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004966202176718723,
      "loss": 5.7263,
      "mean_token_accuracy": 0.15225352942943574,
      "num_tokens": 14623860.0,
      "step": 7035
    },
    {
      "entropy": 5.855080842971802,
      "epoch": 0.6386066763425254,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004966146282152959,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1527144968509674,
      "num_tokens": 14633481.0,
      "step": 7040
    },
    {
      "entropy": 5.809521245956421,
      "epoch": 0.6390602322206096,
      "grad_norm": 2.5,
      "learning_rate": 0.0004966090341756726,
      "loss": 5.6193,
      "mean_token_accuracy": 0.16110151559114455,
      "num_tokens": 14643519.0,
      "step": 7045
    },
    {
      "entropy": 5.722095584869384,
      "epoch": 0.6395137880986937,
      "grad_norm": 2.75,
      "learning_rate": 0.000496603435553118,
      "loss": 5.6306,
      "mean_token_accuracy": 0.16035314351320268,
      "num_tokens": 14654570.0,
      "step": 7050
    },
    {
      "entropy": 5.759281158447266,
      "epoch": 0.6399673439767779,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004965978323477478,
      "loss": 5.6755,
      "mean_token_accuracy": 0.14542552083730698,
      "num_tokens": 14665596.0,
      "step": 7055
    },
    {
      "entropy": 5.8258264541625975,
      "epoch": 0.6404208998548622,
      "grad_norm": 2.78125,
      "learning_rate": 0.000496592224559678,
      "loss": 5.6142,
      "mean_token_accuracy": 0.1545528292655945,
      "num_tokens": 14675660.0,
      "step": 7060
    },
    {
      "entropy": 5.889820384979248,
      "epoch": 0.6408744557329463,
      "grad_norm": 3.25,
      "learning_rate": 0.0004965866121890245,
      "loss": 5.7523,
      "mean_token_accuracy": 0.1493598237633705,
      "num_tokens": 14685229.0,
      "step": 7065
    },
    {
      "entropy": 5.895591783523559,
      "epoch": 0.6413280116110305,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004965809952359034,
      "loss": 5.6761,
      "mean_token_accuracy": 0.1506911963224411,
      "num_tokens": 14695672.0,
      "step": 7070
    },
    {
      "entropy": 5.819560384750366,
      "epoch": 0.6417815674891146,
      "grad_norm": 2.375,
      "learning_rate": 0.0004965753737004309,
      "loss": 5.6134,
      "mean_token_accuracy": 0.15645401328802108,
      "num_tokens": 14705160.0,
      "step": 7075
    },
    {
      "entropy": 5.791318702697754,
      "epoch": 0.6422351233671988,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004965697475827231,
      "loss": 5.6414,
      "mean_token_accuracy": 0.1544739320874214,
      "num_tokens": 14715487.0,
      "step": 7080
    },
    {
      "entropy": 5.8290979862213135,
      "epoch": 0.6426886792452831,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004965641168828965,
      "loss": 5.685,
      "mean_token_accuracy": 0.15276286453008653,
      "num_tokens": 14726987.0,
      "step": 7085
    },
    {
      "entropy": 5.785283803939819,
      "epoch": 0.6431422351233672,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004965584816010676,
      "loss": 5.6615,
      "mean_token_accuracy": 0.15806744545698165,
      "num_tokens": 14737810.0,
      "step": 7090
    },
    {
      "entropy": 5.8629076957702635,
      "epoch": 0.6435957910014514,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004965528417373528,
      "loss": 5.6796,
      "mean_token_accuracy": 0.1506367415189743,
      "num_tokens": 14748997.0,
      "step": 7095
    },
    {
      "entropy": 5.868148517608643,
      "epoch": 0.6440493468795355,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004965471972918688,
      "loss": 5.6911,
      "mean_token_accuracy": 0.15026735216379167,
      "num_tokens": 14760439.0,
      "step": 7100
    },
    {
      "entropy": 5.735028457641602,
      "epoch": 0.6445029027576198,
      "grad_norm": 2.625,
      "learning_rate": 0.0004965415482647322,
      "loss": 5.6046,
      "mean_token_accuracy": 0.15349232405424118,
      "num_tokens": 14770828.0,
      "step": 7105
    },
    {
      "entropy": 5.755033874511719,
      "epoch": 0.6449564586357039,
      "grad_norm": 2.328125,
      "learning_rate": 0.00049653589465606,
      "loss": 5.7486,
      "mean_token_accuracy": 0.15289024412631988,
      "num_tokens": 14781753.0,
      "step": 7110
    },
    {
      "entropy": 5.863271188735962,
      "epoch": 0.6454100145137881,
      "grad_norm": 2.234375,
      "learning_rate": 0.000496530236465969,
      "loss": 5.7505,
      "mean_token_accuracy": 0.14865993708372116,
      "num_tokens": 14792241.0,
      "step": 7115
    },
    {
      "entropy": 5.7898180961608885,
      "epoch": 0.6458635703918723,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004965245736945763,
      "loss": 5.5854,
      "mean_token_accuracy": 0.1545030027627945,
      "num_tokens": 14802580.0,
      "step": 7120
    },
    {
      "entropy": 5.820276069641113,
      "epoch": 0.6463171262699564,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004965189063419989,
      "loss": 5.7744,
      "mean_token_accuracy": 0.151085764169693,
      "num_tokens": 14813103.0,
      "step": 7125
    },
    {
      "entropy": 5.845597743988037,
      "epoch": 0.6467706821480407,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004965132344083541,
      "loss": 5.6569,
      "mean_token_accuracy": 0.15386222600936889,
      "num_tokens": 14823334.0,
      "step": 7130
    },
    {
      "entropy": 5.861242008209229,
      "epoch": 0.6472242380261248,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004965075578937593,
      "loss": 5.6942,
      "mean_token_accuracy": 0.14701148271560668,
      "num_tokens": 14834850.0,
      "step": 7135
    },
    {
      "entropy": 5.782326364517212,
      "epoch": 0.647677793904209,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004965018767983316,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1517026126384735,
      "num_tokens": 14844276.0,
      "step": 7140
    },
    {
      "entropy": 5.726510858535766,
      "epoch": 0.6481313497822931,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004964961911221886,
      "loss": 5.5679,
      "mean_token_accuracy": 0.16671131700277328,
      "num_tokens": 14854352.0,
      "step": 7145
    },
    {
      "entropy": 5.780787515640259,
      "epoch": 0.6485849056603774,
      "grad_norm": 2.609375,
      "learning_rate": 0.000496490500865448,
      "loss": 5.7241,
      "mean_token_accuracy": 0.15255769789218904,
      "num_tokens": 14864395.0,
      "step": 7150
    },
    {
      "entropy": 5.805046892166137,
      "epoch": 0.6490384615384616,
      "grad_norm": 3.375,
      "learning_rate": 0.0004964848060282275,
      "loss": 5.6223,
      "mean_token_accuracy": 0.15498533099889755,
      "num_tokens": 14874373.0,
      "step": 7155
    },
    {
      "entropy": 5.788227224349976,
      "epoch": 0.6494920174165457,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004964791066106447,
      "loss": 5.6011,
      "mean_token_accuracy": 0.1550254613161087,
      "num_tokens": 14884932.0,
      "step": 7160
    },
    {
      "entropy": 5.826607799530029,
      "epoch": 0.6499455732946299,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004964734026128175,
      "loss": 5.7555,
      "mean_token_accuracy": 0.14865583181381226,
      "num_tokens": 14896254.0,
      "step": 7165
    },
    {
      "entropy": 5.9093794345855715,
      "epoch": 0.650399129172714,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004964676940348638,
      "loss": 5.6961,
      "mean_token_accuracy": 0.1514290004968643,
      "num_tokens": 14907392.0,
      "step": 7170
    },
    {
      "entropy": 5.843709802627563,
      "epoch": 0.6508526850507983,
      "grad_norm": 2.40625,
      "learning_rate": 0.000496461980876902,
      "loss": 5.8001,
      "mean_token_accuracy": 0.14915761500597,
      "num_tokens": 14917676.0,
      "step": 7175
    },
    {
      "entropy": 5.839835214614868,
      "epoch": 0.6513062409288825,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004964562631390498,
      "loss": 5.6594,
      "mean_token_accuracy": 0.1541726604104042,
      "num_tokens": 14928500.0,
      "step": 7180
    },
    {
      "entropy": 5.849341201782226,
      "epoch": 0.6517597968069666,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004964505408214257,
      "loss": 5.6167,
      "mean_token_accuracy": 0.15815220177173614,
      "num_tokens": 14938970.0,
      "step": 7185
    },
    {
      "entropy": 5.796083402633667,
      "epoch": 0.6522133526850508,
      "grad_norm": 2.515625,
      "learning_rate": 0.000496444813924148,
      "loss": 5.6739,
      "mean_token_accuracy": 0.15429000556468964,
      "num_tokens": 14949017.0,
      "step": 7190
    },
    {
      "entropy": 5.80718240737915,
      "epoch": 0.652666908563135,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004964390824473351,
      "loss": 5.6101,
      "mean_token_accuracy": 0.1598324030637741,
      "num_tokens": 14958529.0,
      "step": 7195
    },
    {
      "entropy": 5.796051073074341,
      "epoch": 0.6531204644412192,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004964333463911056,
      "loss": 5.6836,
      "mean_token_accuracy": 0.15045272409915925,
      "num_tokens": 14968044.0,
      "step": 7200
    },
    {
      "entropy": 5.83974404335022,
      "epoch": 0.6535740203193033,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004964276057555779,
      "loss": 5.7166,
      "mean_token_accuracy": 0.1514724761247635,
      "num_tokens": 14977867.0,
      "step": 7205
    },
    {
      "entropy": 5.976543140411377,
      "epoch": 0.6540275761973875,
      "grad_norm": 2.109375,
      "learning_rate": 0.000496421860540871,
      "loss": 5.7153,
      "mean_token_accuracy": 0.15487430840730668,
      "num_tokens": 14986936.0,
      "step": 7210
    },
    {
      "entropy": 5.7960106372833256,
      "epoch": 0.6544811320754716,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004964161107471036,
      "loss": 5.6711,
      "mean_token_accuracy": 0.1544204205274582,
      "num_tokens": 14996681.0,
      "step": 7215
    },
    {
      "entropy": 5.7676044464111325,
      "epoch": 0.6549346879535559,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004964103563743946,
      "loss": 5.682,
      "mean_token_accuracy": 0.15293529480695725,
      "num_tokens": 15007100.0,
      "step": 7220
    },
    {
      "entropy": 5.930917358398437,
      "epoch": 0.6553882438316401,
      "grad_norm": 1.875,
      "learning_rate": 0.000496404597422863,
      "loss": 5.7601,
      "mean_token_accuracy": 0.15238583236932754,
      "num_tokens": 15017452.0,
      "step": 7225
    },
    {
      "entropy": 5.838077640533447,
      "epoch": 0.6558417997097242,
      "grad_norm": 2.640625,
      "learning_rate": 0.000496398833892628,
      "loss": 5.6059,
      "mean_token_accuracy": 0.15244304686784743,
      "num_tokens": 15028362.0,
      "step": 7230
    },
    {
      "entropy": 5.829388523101807,
      "epoch": 0.6562953555878084,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004963930657838086,
      "loss": 5.7011,
      "mean_token_accuracy": 0.14786461889743804,
      "num_tokens": 15038691.0,
      "step": 7235
    },
    {
      "entropy": 5.791590785980224,
      "epoch": 0.6567489114658926,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004963872930965242,
      "loss": 5.6779,
      "mean_token_accuracy": 0.15659743323922157,
      "num_tokens": 15050037.0,
      "step": 7240
    },
    {
      "entropy": 5.781030893325806,
      "epoch": 0.6572024673439768,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004963815158308941,
      "loss": 5.63,
      "mean_token_accuracy": 0.15438777953386307,
      "num_tokens": 15059793.0,
      "step": 7245
    },
    {
      "entropy": 5.81955099105835,
      "epoch": 0.657656023222061,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004963757339870379,
      "loss": 5.7275,
      "mean_token_accuracy": 0.149324406683445,
      "num_tokens": 15070500.0,
      "step": 7250
    },
    {
      "entropy": 5.926506376266479,
      "epoch": 0.6581095791001451,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004963699475650751,
      "loss": 5.7868,
      "mean_token_accuracy": 0.14556344151496886,
      "num_tokens": 15081109.0,
      "step": 7255
    },
    {
      "entropy": 5.816445398330688,
      "epoch": 0.6585631349782293,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004963641565651255,
      "loss": 5.6792,
      "mean_token_accuracy": 0.15753332376480103,
      "num_tokens": 15092315.0,
      "step": 7260
    },
    {
      "entropy": 5.796482992172241,
      "epoch": 0.6590166908563135,
      "grad_norm": 2.5,
      "learning_rate": 0.0004963583609873084,
      "loss": 5.7455,
      "mean_token_accuracy": 0.14933345466852188,
      "num_tokens": 15103957.0,
      "step": 7265
    },
    {
      "entropy": 5.874239158630371,
      "epoch": 0.6594702467343977,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004963525608317443,
      "loss": 5.7189,
      "mean_token_accuracy": 0.15017946064472198,
      "num_tokens": 15114996.0,
      "step": 7270
    },
    {
      "entropy": 5.8762486457824705,
      "epoch": 0.6599238026124818,
      "grad_norm": 2.25,
      "learning_rate": 0.0004963467560985527,
      "loss": 5.6769,
      "mean_token_accuracy": 0.14785903841257095,
      "num_tokens": 15125098.0,
      "step": 7275
    },
    {
      "entropy": 5.787175512313842,
      "epoch": 0.660377358490566,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004963409467878538,
      "loss": 5.6366,
      "mean_token_accuracy": 0.1504684343934059,
      "num_tokens": 15134746.0,
      "step": 7280
    },
    {
      "entropy": 5.808853626251221,
      "epoch": 0.6608309143686503,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004963351328997678,
      "loss": 5.6695,
      "mean_token_accuracy": 0.1455580174922943,
      "num_tokens": 15145698.0,
      "step": 7285
    },
    {
      "entropy": 5.806615495681763,
      "epoch": 0.6612844702467344,
      "grad_norm": 6.875,
      "learning_rate": 0.0004963293144344148,
      "loss": 5.7427,
      "mean_token_accuracy": 0.14489729702472687,
      "num_tokens": 15156026.0,
      "step": 7290
    },
    {
      "entropy": 5.85854206085205,
      "epoch": 0.6617380261248186,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004963234913919151,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1468849129974842,
      "num_tokens": 15165900.0,
      "step": 7295
    },
    {
      "entropy": 5.836259555816651,
      "epoch": 0.6621915820029027,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004963176637723893,
      "loss": 5.7369,
      "mean_token_accuracy": 0.14796246439218522,
      "num_tokens": 15176693.0,
      "step": 7300
    },
    {
      "entropy": 5.850326919555664,
      "epoch": 0.6626451378809869,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004963118315759578,
      "loss": 5.7717,
      "mean_token_accuracy": 0.14276785105466844,
      "num_tokens": 15186535.0,
      "step": 7305
    },
    {
      "entropy": 5.898229312896729,
      "epoch": 0.6630986937590712,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004963059948027411,
      "loss": 5.6684,
      "mean_token_accuracy": 0.14483277201652528,
      "num_tokens": 15196019.0,
      "step": 7310
    },
    {
      "entropy": 5.776585054397583,
      "epoch": 0.6635522496371553,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004963001534528601,
      "loss": 5.5405,
      "mean_token_accuracy": 0.1623222604393959,
      "num_tokens": 15205273.0,
      "step": 7315
    },
    {
      "entropy": 5.7897053241729735,
      "epoch": 0.6640058055152395,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004962943075264355,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1488634929060936,
      "num_tokens": 15216716.0,
      "step": 7320
    },
    {
      "entropy": 5.807294750213623,
      "epoch": 0.6644593613933236,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004962884570235884,
      "loss": 5.7236,
      "mean_token_accuracy": 0.15068677812814713,
      "num_tokens": 15227306.0,
      "step": 7325
    },
    {
      "entropy": 5.742990303039551,
      "epoch": 0.6649129172714079,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004962826019444394,
      "loss": 5.599,
      "mean_token_accuracy": 0.1591808944940567,
      "num_tokens": 15237302.0,
      "step": 7330
    },
    {
      "entropy": 5.827797937393188,
      "epoch": 0.665366473149492,
      "grad_norm": 2.25,
      "learning_rate": 0.0004962767422891099,
      "loss": 5.6475,
      "mean_token_accuracy": 0.15172240287065505,
      "num_tokens": 15247064.0,
      "step": 7335
    },
    {
      "entropy": 5.857191896438598,
      "epoch": 0.6658200290275762,
      "grad_norm": 3.078125,
      "learning_rate": 0.000496270878057721,
      "loss": 5.7054,
      "mean_token_accuracy": 0.15546690672636032,
      "num_tokens": 15257006.0,
      "step": 7340
    },
    {
      "entropy": 5.8285126209259035,
      "epoch": 0.6662735849056604,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004962650092503939,
      "loss": 5.683,
      "mean_token_accuracy": 0.15548086911439896,
      "num_tokens": 15266720.0,
      "step": 7345
    },
    {
      "entropy": 5.79559907913208,
      "epoch": 0.6667271407837445,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004962591358672502,
      "loss": 5.6971,
      "mean_token_accuracy": 0.150240645557642,
      "num_tokens": 15277330.0,
      "step": 7350
    },
    {
      "entropy": 5.8351600646972654,
      "epoch": 0.6671806966618288,
      "grad_norm": 7.84375,
      "learning_rate": 0.000496253257908411,
      "loss": 5.6504,
      "mean_token_accuracy": 0.15267718434333802,
      "num_tokens": 15287401.0,
      "step": 7355
    },
    {
      "entropy": 5.7537617683410645,
      "epoch": 0.6676342525399129,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004962473753739981,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1605018511414528,
      "num_tokens": 15297426.0,
      "step": 7360
    },
    {
      "entropy": 5.8254279613494875,
      "epoch": 0.6680878084179971,
      "grad_norm": 2.578125,
      "learning_rate": 0.000496241488264133,
      "loss": 5.5919,
      "mean_token_accuracy": 0.15262950733304023,
      "num_tokens": 15308412.0,
      "step": 7365
    },
    {
      "entropy": 5.769833183288574,
      "epoch": 0.6685413642960812,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004962355965789377,
      "loss": 5.6122,
      "mean_token_accuracy": 0.14959425926208497,
      "num_tokens": 15318821.0,
      "step": 7370
    },
    {
      "entropy": 5.800112342834472,
      "epoch": 0.6689949201741655,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004962297003185338,
      "loss": 5.6561,
      "mean_token_accuracy": 0.1527245745062828,
      "num_tokens": 15328737.0,
      "step": 7375
    },
    {
      "entropy": 5.918911552429199,
      "epoch": 0.6694484760522497,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004962237994830433,
      "loss": 5.7712,
      "mean_token_accuracy": 0.14499044567346572,
      "num_tokens": 15339695.0,
      "step": 7380
    },
    {
      "entropy": 5.899818372726441,
      "epoch": 0.6699020319303338,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004962178940725883,
      "loss": 5.7177,
      "mean_token_accuracy": 0.14840016067028045,
      "num_tokens": 15350270.0,
      "step": 7385
    },
    {
      "entropy": 5.827748537063599,
      "epoch": 0.670355587808418,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004962119840872908,
      "loss": 5.7215,
      "mean_token_accuracy": 0.15222820043563842,
      "num_tokens": 15360986.0,
      "step": 7390
    },
    {
      "entropy": 5.822416067123413,
      "epoch": 0.6708091436865021,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004962060695272733,
      "loss": 5.7284,
      "mean_token_accuracy": 0.14872562736272812,
      "num_tokens": 15371786.0,
      "step": 7395
    },
    {
      "entropy": 5.847713041305542,
      "epoch": 0.6712626995645864,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004962001503926577,
      "loss": 5.664,
      "mean_token_accuracy": 0.15866391956806183,
      "num_tokens": 15382960.0,
      "step": 7400
    },
    {
      "entropy": 5.81244649887085,
      "epoch": 0.6717162554426706,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004961942266835667,
      "loss": 5.6593,
      "mean_token_accuracy": 0.15917277187108994,
      "num_tokens": 15392807.0,
      "step": 7405
    },
    {
      "entropy": 5.838684129714966,
      "epoch": 0.6721698113207547,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004961882984001227,
      "loss": 5.7278,
      "mean_token_accuracy": 0.14638093262910842,
      "num_tokens": 15403712.0,
      "step": 7410
    },
    {
      "entropy": 5.842763519287109,
      "epoch": 0.6726233671988389,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004961823655424484,
      "loss": 5.7594,
      "mean_token_accuracy": 0.14760402739048004,
      "num_tokens": 15414839.0,
      "step": 7415
    },
    {
      "entropy": 5.85579662322998,
      "epoch": 0.6730769230769231,
      "grad_norm": 3.375,
      "learning_rate": 0.0004961764281106664,
      "loss": 5.6555,
      "mean_token_accuracy": 0.15158698484301567,
      "num_tokens": 15425320.0,
      "step": 7420
    },
    {
      "entropy": 5.880425500869751,
      "epoch": 0.6735304789550073,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004961704861048995,
      "loss": 5.6446,
      "mean_token_accuracy": 0.15110456049442292,
      "num_tokens": 15435476.0,
      "step": 7425
    },
    {
      "entropy": 5.793002033233643,
      "epoch": 0.6739840348330914,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004961645395252705,
      "loss": 5.7385,
      "mean_token_accuracy": 0.14751248359680175,
      "num_tokens": 15445786.0,
      "step": 7430
    },
    {
      "entropy": 5.762565755844117,
      "epoch": 0.6744375907111756,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004961585883719025,
      "loss": 5.7483,
      "mean_token_accuracy": 0.152704431861639,
      "num_tokens": 15455915.0,
      "step": 7435
    },
    {
      "entropy": 5.846822690963745,
      "epoch": 0.6748911465892597,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004961526326449186,
      "loss": 5.7176,
      "mean_token_accuracy": 0.15189335495233536,
      "num_tokens": 15466523.0,
      "step": 7440
    },
    {
      "entropy": 5.879119920730591,
      "epoch": 0.675344702467344,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004961466723444419,
      "loss": 5.7177,
      "mean_token_accuracy": 0.1526683509349823,
      "num_tokens": 15476821.0,
      "step": 7445
    },
    {
      "entropy": 5.80903730392456,
      "epoch": 0.6757982583454282,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004961407074705955,
      "loss": 5.6795,
      "mean_token_accuracy": 0.15011930987238883,
      "num_tokens": 15487182.0,
      "step": 7450
    },
    {
      "entropy": 5.812462520599365,
      "epoch": 0.6762518142235123,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004961347380235031,
      "loss": 5.6465,
      "mean_token_accuracy": 0.15409299433231355,
      "num_tokens": 15497600.0,
      "step": 7455
    },
    {
      "entropy": 5.885445642471313,
      "epoch": 0.6767053701015965,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004961287640032879,
      "loss": 5.7799,
      "mean_token_accuracy": 0.14085702300071717,
      "num_tokens": 15508278.0,
      "step": 7460
    },
    {
      "entropy": 5.931916284561157,
      "epoch": 0.6771589259796807,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004961227854100734,
      "loss": 5.8092,
      "mean_token_accuracy": 0.14859505146741867,
      "num_tokens": 15518458.0,
      "step": 7465
    },
    {
      "entropy": 5.752091789245606,
      "epoch": 0.6776124818577649,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004961168022439835,
      "loss": 5.6075,
      "mean_token_accuracy": 0.15934648662805556,
      "num_tokens": 15528738.0,
      "step": 7470
    },
    {
      "entropy": 5.784538269042969,
      "epoch": 0.6780660377358491,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004961108145051417,
      "loss": 5.6542,
      "mean_token_accuracy": 0.15490319281816484,
      "num_tokens": 15538656.0,
      "step": 7475
    },
    {
      "entropy": 5.7982035160064695,
      "epoch": 0.6785195936139332,
      "grad_norm": 3.515625,
      "learning_rate": 0.000496104822193672,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1550088033080101,
      "num_tokens": 15549285.0,
      "step": 7480
    },
    {
      "entropy": 5.904118776321411,
      "epoch": 0.6789731494920174,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004960988253096982,
      "loss": 5.7249,
      "mean_token_accuracy": 0.15838196128606796,
      "num_tokens": 15559766.0,
      "step": 7485
    },
    {
      "entropy": 5.829934310913086,
      "epoch": 0.6794267053701016,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004960928238533443,
      "loss": 5.6062,
      "mean_token_accuracy": 0.15969916731119155,
      "num_tokens": 15570873.0,
      "step": 7490
    },
    {
      "entropy": 5.828317451477051,
      "epoch": 0.6798802612481858,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004960868178247344,
      "loss": 5.8395,
      "mean_token_accuracy": 0.1505518451333046,
      "num_tokens": 15581229.0,
      "step": 7495
    },
    {
      "entropy": 5.747046995162964,
      "epoch": 0.68033381712627,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004960808072239929,
      "loss": 5.7085,
      "mean_token_accuracy": 0.1462366871535778,
      "num_tokens": 15591686.0,
      "step": 7500
    },
    {
      "entropy": 5.912446784973144,
      "epoch": 0.6807873730043541,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004960747920512439,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1496862567961216,
      "num_tokens": 15601869.0,
      "step": 7505
    },
    {
      "entropy": 5.829827451705933,
      "epoch": 0.6812409288824384,
      "grad_norm": 2.234375,
      "learning_rate": 0.000496068772306612,
      "loss": 5.7228,
      "mean_token_accuracy": 0.14869892299175264,
      "num_tokens": 15612946.0,
      "step": 7510
    },
    {
      "entropy": 5.813806962966919,
      "epoch": 0.6816944847605225,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004960627479902214,
      "loss": 5.7616,
      "mean_token_accuracy": 0.14657169133424758,
      "num_tokens": 15623761.0,
      "step": 7515
    },
    {
      "entropy": 5.797790241241455,
      "epoch": 0.6821480406386067,
      "grad_norm": 3.453125,
      "learning_rate": 0.000496056719102197,
      "loss": 5.7404,
      "mean_token_accuracy": 0.14438045471906663,
      "num_tokens": 15634500.0,
      "step": 7520
    },
    {
      "entropy": 5.746096229553222,
      "epoch": 0.6826015965166908,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004960506856426631,
      "loss": 5.6124,
      "mean_token_accuracy": 0.149461767077446,
      "num_tokens": 15644896.0,
      "step": 7525
    },
    {
      "entropy": 5.8980224609375,
      "epoch": 0.683055152394775,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004960446476117448,
      "loss": 5.7132,
      "mean_token_accuracy": 0.15735317766666412,
      "num_tokens": 15655008.0,
      "step": 7530
    },
    {
      "entropy": 5.8324168682098385,
      "epoch": 0.6835087082728593,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004960386050095668,
      "loss": 5.6322,
      "mean_token_accuracy": 0.14784073680639268,
      "num_tokens": 15665944.0,
      "step": 7535
    },
    {
      "entropy": 5.824177026748657,
      "epoch": 0.6839622641509434,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004960325578362543,
      "loss": 5.7564,
      "mean_token_accuracy": 0.15302710086107255,
      "num_tokens": 15677099.0,
      "step": 7540
    },
    {
      "entropy": 5.838755798339844,
      "epoch": 0.6844158200290276,
      "grad_norm": 3.109375,
      "learning_rate": 0.000496026506091932,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1540417954325676,
      "num_tokens": 15687531.0,
      "step": 7545
    },
    {
      "entropy": 5.907812118530273,
      "epoch": 0.6848693759071117,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004960204497767253,
      "loss": 5.7004,
      "mean_token_accuracy": 0.14849745184183122,
      "num_tokens": 15697002.0,
      "step": 7550
    },
    {
      "entropy": 5.784596347808838,
      "epoch": 0.685322931785196,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004960143888907594,
      "loss": 5.6055,
      "mean_token_accuracy": 0.14923751652240752,
      "num_tokens": 15707249.0,
      "step": 7555
    },
    {
      "entropy": 5.801516056060791,
      "epoch": 0.6857764876632801,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004960083234341595,
      "loss": 5.6647,
      "mean_token_accuracy": 0.15030739307403565,
      "num_tokens": 15718764.0,
      "step": 7560
    },
    {
      "entropy": 5.818806838989258,
      "epoch": 0.6862300435413643,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004960022534070513,
      "loss": 5.6246,
      "mean_token_accuracy": 0.15101515799760817,
      "num_tokens": 15728854.0,
      "step": 7565
    },
    {
      "entropy": 5.813277912139893,
      "epoch": 0.6866835994194485,
      "grad_norm": 2.125,
      "learning_rate": 0.00049599617880956,
      "loss": 5.6366,
      "mean_token_accuracy": 0.15327175408601762,
      "num_tokens": 15739682.0,
      "step": 7570
    },
    {
      "entropy": 5.815754461288452,
      "epoch": 0.6871371552975326,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004959900996418115,
      "loss": 5.6278,
      "mean_token_accuracy": 0.16303820610046388,
      "num_tokens": 15749628.0,
      "step": 7575
    },
    {
      "entropy": 5.768430137634278,
      "epoch": 0.6875907111756169,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004959840159039315,
      "loss": 5.7055,
      "mean_token_accuracy": 0.15597921907901763,
      "num_tokens": 15760137.0,
      "step": 7580
    },
    {
      "entropy": 5.7924089431762695,
      "epoch": 0.688044267053701,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004959779275960456,
      "loss": 5.6656,
      "mean_token_accuracy": 0.150012369453907,
      "num_tokens": 15769889.0,
      "step": 7585
    },
    {
      "entropy": 5.846169757843017,
      "epoch": 0.6884978229317852,
      "grad_norm": 5.4375,
      "learning_rate": 0.00049597183471828,
      "loss": 5.7515,
      "mean_token_accuracy": 0.15068214386701584,
      "num_tokens": 15780912.0,
      "step": 7590
    },
    {
      "entropy": 5.89394326210022,
      "epoch": 0.6889513788098693,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004959657372707604,
      "loss": 5.634,
      "mean_token_accuracy": 0.15749605894088745,
      "num_tokens": 15791719.0,
      "step": 7595
    },
    {
      "entropy": 5.874384260177612,
      "epoch": 0.6894049346879536,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004959596352536132,
      "loss": 5.6705,
      "mean_token_accuracy": 0.15589183270931245,
      "num_tokens": 15802551.0,
      "step": 7600
    },
    {
      "entropy": 5.864084911346436,
      "epoch": 0.6898584905660378,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004959535286669643,
      "loss": 5.7224,
      "mean_token_accuracy": 0.15257749259471892,
      "num_tokens": 15813551.0,
      "step": 7605
    },
    {
      "entropy": 5.830462169647217,
      "epoch": 0.6903120464441219,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004959474175109402,
      "loss": 5.6989,
      "mean_token_accuracy": 0.15375184565782546,
      "num_tokens": 15824262.0,
      "step": 7610
    },
    {
      "entropy": 5.864915943145752,
      "epoch": 0.6907656023222061,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004959413017856672,
      "loss": 5.7468,
      "mean_token_accuracy": 0.1529516577720642,
      "num_tokens": 15835493.0,
      "step": 7615
    },
    {
      "entropy": 5.750688648223877,
      "epoch": 0.6912191582002902,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004959351814912716,
      "loss": 5.6216,
      "mean_token_accuracy": 0.15190897583961488,
      "num_tokens": 15846224.0,
      "step": 7620
    },
    {
      "entropy": 5.812427377700805,
      "epoch": 0.6916727140783745,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004959290566278803,
      "loss": 5.6666,
      "mean_token_accuracy": 0.15235227942466736,
      "num_tokens": 15855622.0,
      "step": 7625
    },
    {
      "entropy": 5.867595767974853,
      "epoch": 0.6921262699564587,
      "grad_norm": 2.25,
      "learning_rate": 0.0004959229271956199,
      "loss": 5.7211,
      "mean_token_accuracy": 0.15340991020202638,
      "num_tokens": 15865835.0,
      "step": 7630
    },
    {
      "entropy": 5.828436231613159,
      "epoch": 0.6925798258345428,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004959167931946169,
      "loss": 5.6652,
      "mean_token_accuracy": 0.14981707483530043,
      "num_tokens": 15875915.0,
      "step": 7635
    },
    {
      "entropy": 5.883977460861206,
      "epoch": 0.693033381712627,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004959106546249984,
      "loss": 5.6936,
      "mean_token_accuracy": 0.1512582406401634,
      "num_tokens": 15887111.0,
      "step": 7640
    },
    {
      "entropy": 5.899054718017578,
      "epoch": 0.6934869375907112,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004959045114868913,
      "loss": 5.8056,
      "mean_token_accuracy": 0.13988058194518088,
      "num_tokens": 15899528.0,
      "step": 7645
    },
    {
      "entropy": 5.864440107345581,
      "epoch": 0.6939404934687954,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004958983637804225,
      "loss": 5.696,
      "mean_token_accuracy": 0.15691021382808684,
      "num_tokens": 15910881.0,
      "step": 7650
    },
    {
      "entropy": 5.844481897354126,
      "epoch": 0.6943940493468795,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004958922115057194,
      "loss": 5.7601,
      "mean_token_accuracy": 0.14802518784999846,
      "num_tokens": 15921189.0,
      "step": 7655
    },
    {
      "entropy": 5.890579032897949,
      "epoch": 0.6948476052249637,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004958860546629089,
      "loss": 5.7344,
      "mean_token_accuracy": 0.15592280477285386,
      "num_tokens": 15931276.0,
      "step": 7660
    },
    {
      "entropy": 5.924163198471069,
      "epoch": 0.6953011611030478,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004958798932521186,
      "loss": 5.8199,
      "mean_token_accuracy": 0.14040146842598916,
      "num_tokens": 15941506.0,
      "step": 7665
    },
    {
      "entropy": 5.763565111160278,
      "epoch": 0.6957547169811321,
      "grad_norm": 2.875,
      "learning_rate": 0.0004958737272734759,
      "loss": 5.639,
      "mean_token_accuracy": 0.15290124863386154,
      "num_tokens": 15951158.0,
      "step": 7670
    },
    {
      "entropy": 5.818205165863037,
      "epoch": 0.6962082728592163,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004958675567271081,
      "loss": 5.6771,
      "mean_token_accuracy": 0.15895491987466812,
      "num_tokens": 15960852.0,
      "step": 7675
    },
    {
      "entropy": 5.8290722370147705,
      "epoch": 0.6966618287373004,
      "grad_norm": 2.609375,
      "learning_rate": 0.000495861381613143,
      "loss": 5.6803,
      "mean_token_accuracy": 0.15305335968732833,
      "num_tokens": 15971291.0,
      "step": 7680
    },
    {
      "entropy": 5.846955585479736,
      "epoch": 0.6971153846153846,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004958552019317082,
      "loss": 5.7539,
      "mean_token_accuracy": 0.1479065515100956,
      "num_tokens": 15982180.0,
      "step": 7685
    },
    {
      "entropy": 5.797716379165649,
      "epoch": 0.6975689404934688,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004958490176829316,
      "loss": 5.6355,
      "mean_token_accuracy": 0.15521086752414703,
      "num_tokens": 15992363.0,
      "step": 7690
    },
    {
      "entropy": 5.861604118347168,
      "epoch": 0.698022496371553,
      "grad_norm": 3.265625,
      "learning_rate": 0.000495842828866941,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1435869313776493,
      "num_tokens": 16003477.0,
      "step": 7695
    },
    {
      "entropy": 5.885314273834228,
      "epoch": 0.6984760522496372,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004958366354838645,
      "loss": 5.6907,
      "mean_token_accuracy": 0.15125567615032195,
      "num_tokens": 16013747.0,
      "step": 7700
    },
    {
      "entropy": 5.860510921478271,
      "epoch": 0.6989296081277213,
      "grad_norm": 4.0,
      "learning_rate": 0.00049583043753383,
      "loss": 5.7163,
      "mean_token_accuracy": 0.1556942120194435,
      "num_tokens": 16025662.0,
      "step": 7705
    },
    {
      "entropy": 5.890257215499878,
      "epoch": 0.6993831640058055,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004958242350169658,
      "loss": 5.7906,
      "mean_token_accuracy": 0.15480666384100913,
      "num_tokens": 16037315.0,
      "step": 7710
    },
    {
      "entropy": 5.741261672973633,
      "epoch": 0.6998367198838897,
      "grad_norm": 2.5,
      "learning_rate": 0.0004958180279334001,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1540132313966751,
      "num_tokens": 16047297.0,
      "step": 7715
    },
    {
      "entropy": 5.783619117736817,
      "epoch": 0.7002902757619739,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004958118162832614,
      "loss": 5.5993,
      "mean_token_accuracy": 0.1512565240263939,
      "num_tokens": 16057573.0,
      "step": 7720
    },
    {
      "entropy": 5.8567568302154545,
      "epoch": 0.700743831640058,
      "grad_norm": 3.0,
      "learning_rate": 0.0004958056000666781,
      "loss": 5.7383,
      "mean_token_accuracy": 0.14643106907606124,
      "num_tokens": 16067926.0,
      "step": 7725
    },
    {
      "entropy": 5.880869722366333,
      "epoch": 0.7011973875181422,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004957993792837786,
      "loss": 5.6846,
      "mean_token_accuracy": 0.14718203395605087,
      "num_tokens": 16077938.0,
      "step": 7730
    },
    {
      "entropy": 5.8295722007751465,
      "epoch": 0.7016509433962265,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004957931539346917,
      "loss": 5.5905,
      "mean_token_accuracy": 0.15416589826345445,
      "num_tokens": 16088163.0,
      "step": 7735
    },
    {
      "entropy": 5.909097576141358,
      "epoch": 0.7021044992743106,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004957869240195462,
      "loss": 5.8017,
      "mean_token_accuracy": 0.14014994204044343,
      "num_tokens": 16100640.0,
      "step": 7740
    },
    {
      "entropy": 5.823917055130005,
      "epoch": 0.7025580551523948,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004957806895384707,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15346488058567048,
      "num_tokens": 16111018.0,
      "step": 7745
    },
    {
      "entropy": 5.7712815284729,
      "epoch": 0.7030116110304789,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004957744504915944,
      "loss": 5.6513,
      "mean_token_accuracy": 0.15386599451303482,
      "num_tokens": 16120873.0,
      "step": 7750
    },
    {
      "entropy": 5.8102219581604,
      "epoch": 0.7034651669085631,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004957682068790463,
      "loss": 5.666,
      "mean_token_accuracy": 0.15427257865667343,
      "num_tokens": 16131178.0,
      "step": 7755
    },
    {
      "entropy": 5.777589178085327,
      "epoch": 0.7039187227866474,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004957619587009553,
      "loss": 5.662,
      "mean_token_accuracy": 0.15195267498493195,
      "num_tokens": 16141156.0,
      "step": 7760
    },
    {
      "entropy": 5.823400831222534,
      "epoch": 0.7043722786647315,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004957557059574507,
      "loss": 5.7764,
      "mean_token_accuracy": 0.14289068877696992,
      "num_tokens": 16150077.0,
      "step": 7765
    },
    {
      "entropy": 5.917142295837403,
      "epoch": 0.7048258345428157,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004957494486486618,
      "loss": 5.6877,
      "mean_token_accuracy": 0.15153144299983978,
      "num_tokens": 16160514.0,
      "step": 7770
    },
    {
      "entropy": 5.9026023864746096,
      "epoch": 0.7052793904208998,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004957431867747182,
      "loss": 5.7275,
      "mean_token_accuracy": 0.15452097058296205,
      "num_tokens": 16170811.0,
      "step": 7775
    },
    {
      "entropy": 5.7668298244476315,
      "epoch": 0.7057329462989841,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004957369203357492,
      "loss": 5.6154,
      "mean_token_accuracy": 0.15205578207969667,
      "num_tokens": 16181553.0,
      "step": 7780
    },
    {
      "entropy": 5.758222198486328,
      "epoch": 0.7061865021770682,
      "grad_norm": 2.25,
      "learning_rate": 0.0004957306493318844,
      "loss": 5.6736,
      "mean_token_accuracy": 0.15890712887048722,
      "num_tokens": 16191308.0,
      "step": 7785
    },
    {
      "entropy": 5.816359376907348,
      "epoch": 0.7066400580551524,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004957243737632535,
      "loss": 5.6282,
      "mean_token_accuracy": 0.15420712679624557,
      "num_tokens": 16201311.0,
      "step": 7790
    },
    {
      "entropy": 5.890701484680176,
      "epoch": 0.7070936139332366,
      "grad_norm": 3.125,
      "learning_rate": 0.0004957180936299863,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1514902353286743,
      "num_tokens": 16211105.0,
      "step": 7795
    },
    {
      "entropy": 5.8345139503479,
      "epoch": 0.7075471698113207,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004957118089322127,
      "loss": 5.6679,
      "mean_token_accuracy": 0.14869241565465927,
      "num_tokens": 16220836.0,
      "step": 7800
    },
    {
      "entropy": 5.7877466678619385,
      "epoch": 0.708000725689405,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004957055196700628,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1567522630095482,
      "num_tokens": 16231630.0,
      "step": 7805
    },
    {
      "entropy": 5.844862079620361,
      "epoch": 0.7084542815674891,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004956992258436663,
      "loss": 5.7023,
      "mean_token_accuracy": 0.1523435965180397,
      "num_tokens": 16242272.0,
      "step": 7810
    },
    {
      "entropy": 5.744817018508911,
      "epoch": 0.7089078374455733,
      "grad_norm": 5.3125,
      "learning_rate": 0.0004956929274531536,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1529250368475914,
      "num_tokens": 16252555.0,
      "step": 7815
    },
    {
      "entropy": 5.78554220199585,
      "epoch": 0.7093613933236574,
      "grad_norm": 2.9375,
      "learning_rate": 0.000495686624498655,
      "loss": 5.72,
      "mean_token_accuracy": 0.15313845574855806,
      "num_tokens": 16261964.0,
      "step": 7820
    },
    {
      "entropy": 5.9008283615112305,
      "epoch": 0.7098149492017417,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004956803169803007,
      "loss": 5.778,
      "mean_token_accuracy": 0.14940793365240096,
      "num_tokens": 16274543.0,
      "step": 7825
    },
    {
      "entropy": 5.838950538635254,
      "epoch": 0.7102685050798259,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004956740048982212,
      "loss": 5.6871,
      "mean_token_accuracy": 0.14389506876468658,
      "num_tokens": 16285124.0,
      "step": 7830
    },
    {
      "entropy": 5.811396932601928,
      "epoch": 0.71072206095791,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004956676882525472,
      "loss": 5.7509,
      "mean_token_accuracy": 0.15540383011102676,
      "num_tokens": 16295727.0,
      "step": 7835
    },
    {
      "entropy": 5.8709159851074215,
      "epoch": 0.7111756168359942,
      "grad_norm": 3.234375,
      "learning_rate": 0.000495661367043409,
      "loss": 5.6656,
      "mean_token_accuracy": 0.15600730031728743,
      "num_tokens": 16305496.0,
      "step": 7840
    },
    {
      "entropy": 5.87369556427002,
      "epoch": 0.7116291727140783,
      "grad_norm": 6.0,
      "learning_rate": 0.0004956550412709376,
      "loss": 5.6378,
      "mean_token_accuracy": 0.15536927729845046,
      "num_tokens": 16315112.0,
      "step": 7845
    },
    {
      "entropy": 5.886559772491455,
      "epoch": 0.7120827285921626,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004956487109352637,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1533990427851677,
      "num_tokens": 16325613.0,
      "step": 7850
    },
    {
      "entropy": 5.820042085647583,
      "epoch": 0.7125362844702468,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004956423760365181,
      "loss": 5.6057,
      "mean_token_accuracy": 0.15789132863283156,
      "num_tokens": 16335317.0,
      "step": 7855
    },
    {
      "entropy": 5.794498109817505,
      "epoch": 0.7129898403483309,
      "grad_norm": 7.625,
      "learning_rate": 0.0004956360365748321,
      "loss": 5.7599,
      "mean_token_accuracy": 0.14459077566862105,
      "num_tokens": 16345003.0,
      "step": 7860
    },
    {
      "entropy": 5.861040449142456,
      "epoch": 0.7134433962264151,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004956296925503367,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14798754081130028,
      "num_tokens": 16356108.0,
      "step": 7865
    },
    {
      "entropy": 5.858279132843018,
      "epoch": 0.7138969521044993,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004956233439631628,
      "loss": 5.562,
      "mean_token_accuracy": 0.15614887624979018,
      "num_tokens": 16364733.0,
      "step": 7870
    },
    {
      "entropy": 5.874885559082031,
      "epoch": 0.7143505079825835,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004956169908134421,
      "loss": 5.6638,
      "mean_token_accuracy": 0.15401551127433777,
      "num_tokens": 16375229.0,
      "step": 7875
    },
    {
      "entropy": 5.813601684570313,
      "epoch": 0.7148040638606676,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004956106331013057,
      "loss": 5.756,
      "mean_token_accuracy": 0.15068020671606064,
      "num_tokens": 16386366.0,
      "step": 7880
    },
    {
      "entropy": 5.802809762954712,
      "epoch": 0.7152576197387518,
      "grad_norm": 2.875,
      "learning_rate": 0.0004956042708268853,
      "loss": 5.6342,
      "mean_token_accuracy": 0.15512944683432578,
      "num_tokens": 16397067.0,
      "step": 7885
    },
    {
      "entropy": 5.80576000213623,
      "epoch": 0.715711175616836,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004955979039903123,
      "loss": 5.6023,
      "mean_token_accuracy": 0.15368096828460692,
      "num_tokens": 16407256.0,
      "step": 7890
    },
    {
      "entropy": 5.754089975357056,
      "epoch": 0.7161647314949202,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004955915325917183,
      "loss": 5.6564,
      "mean_token_accuracy": 0.15151944160461425,
      "num_tokens": 16418701.0,
      "step": 7895
    },
    {
      "entropy": 5.782627153396606,
      "epoch": 0.7166182873730044,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004955851566312354,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1544649764895439,
      "num_tokens": 16428019.0,
      "step": 7900
    },
    {
      "entropy": 5.766601467132569,
      "epoch": 0.7170718432510885,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004955787761089952,
      "loss": 5.6641,
      "mean_token_accuracy": 0.1493433177471161,
      "num_tokens": 16440122.0,
      "step": 7905
    },
    {
      "entropy": 5.884334135055542,
      "epoch": 0.7175253991291727,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004955723910251296,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1512959912419319,
      "num_tokens": 16449801.0,
      "step": 7910
    },
    {
      "entropy": 5.7857404232025145,
      "epoch": 0.717978955007257,
      "grad_norm": 3.375,
      "learning_rate": 0.0004955660013797709,
      "loss": 5.5926,
      "mean_token_accuracy": 0.15442865192890168,
      "num_tokens": 16460155.0,
      "step": 7915
    },
    {
      "entropy": 5.845020294189453,
      "epoch": 0.7184325108853411,
      "grad_norm": 2.75,
      "learning_rate": 0.000495559607173051,
      "loss": 5.7866,
      "mean_token_accuracy": 0.14254856258630752,
      "num_tokens": 16470656.0,
      "step": 7920
    },
    {
      "entropy": 5.876619863510132,
      "epoch": 0.7188860667634253,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004955532084051022,
      "loss": 5.6977,
      "mean_token_accuracy": 0.15091280341148378,
      "num_tokens": 16481449.0,
      "step": 7925
    },
    {
      "entropy": 5.9208259105682375,
      "epoch": 0.7193396226415094,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004955468050760568,
      "loss": 5.7113,
      "mean_token_accuracy": 0.15328221768140793,
      "num_tokens": 16492032.0,
      "step": 7930
    },
    {
      "entropy": 5.832501029968261,
      "epoch": 0.7197931785195936,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004955403971860474,
      "loss": 5.6446,
      "mean_token_accuracy": 0.1522648885846138,
      "num_tokens": 16502221.0,
      "step": 7935
    },
    {
      "entropy": 5.8008585453033445,
      "epoch": 0.7202467343976778,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004955339847352063,
      "loss": 5.6355,
      "mean_token_accuracy": 0.15295949429273606,
      "num_tokens": 16512021.0,
      "step": 7940
    },
    {
      "entropy": 5.8326164245605465,
      "epoch": 0.720700290275762,
      "grad_norm": 4.875,
      "learning_rate": 0.0004955275677236662,
      "loss": 5.6849,
      "mean_token_accuracy": 0.14992091357707976,
      "num_tokens": 16522060.0,
      "step": 7945
    },
    {
      "entropy": 5.882563972473145,
      "epoch": 0.7211538461538461,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004955211461515599,
      "loss": 5.6814,
      "mean_token_accuracy": 0.155203914642334,
      "num_tokens": 16532827.0,
      "step": 7950
    },
    {
      "entropy": 5.8585892677307125,
      "epoch": 0.7216074020319303,
      "grad_norm": 3.46875,
      "learning_rate": 0.00049551472001902,
      "loss": 5.7052,
      "mean_token_accuracy": 0.15561985224485397,
      "num_tokens": 16542838.0,
      "step": 7955
    },
    {
      "entropy": 5.779575490951538,
      "epoch": 0.7220609579100146,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004955082893261796,
      "loss": 5.6582,
      "mean_token_accuracy": 0.15778737068176268,
      "num_tokens": 16552615.0,
      "step": 7960
    },
    {
      "entropy": 5.908274412155151,
      "epoch": 0.7225145137880987,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004955018540731714,
      "loss": 5.726,
      "mean_token_accuracy": 0.14718418419361115,
      "num_tokens": 16561777.0,
      "step": 7965
    },
    {
      "entropy": 5.918739891052246,
      "epoch": 0.7229680696661829,
      "grad_norm": 5.125,
      "learning_rate": 0.0004954954142601288,
      "loss": 5.6905,
      "mean_token_accuracy": 0.1526739552617073,
      "num_tokens": 16572111.0,
      "step": 7970
    },
    {
      "entropy": 5.830246782302856,
      "epoch": 0.723421625544267,
      "grad_norm": 1.875,
      "learning_rate": 0.0004954889698871849,
      "loss": 5.7246,
      "mean_token_accuracy": 0.15518662035465242,
      "num_tokens": 16581674.0,
      "step": 7975
    },
    {
      "entropy": 5.847196817398071,
      "epoch": 0.7238751814223512,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004954825209544728,
      "loss": 5.7107,
      "mean_token_accuracy": 0.15622829049825668,
      "num_tokens": 16591802.0,
      "step": 7980
    },
    {
      "entropy": 5.819644355773926,
      "epoch": 0.7243287373004355,
      "grad_norm": 3.90625,
      "learning_rate": 0.000495476067462126,
      "loss": 5.6259,
      "mean_token_accuracy": 0.1592419072985649,
      "num_tokens": 16602772.0,
      "step": 7985
    },
    {
      "entropy": 5.8849729061126705,
      "epoch": 0.7247822931785196,
      "grad_norm": 3.421875,
      "learning_rate": 0.000495469609410278,
      "loss": 5.669,
      "mean_token_accuracy": 0.15404960215091706,
      "num_tokens": 16613526.0,
      "step": 7990
    },
    {
      "entropy": 5.807653331756592,
      "epoch": 0.7252358490566038,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004954631467990622,
      "loss": 5.6769,
      "mean_token_accuracy": 0.1550093710422516,
      "num_tokens": 16623583.0,
      "step": 7995
    },
    {
      "entropy": 5.852053880691528,
      "epoch": 0.7256894049346879,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004954566796286124,
      "loss": 5.7295,
      "mean_token_accuracy": 0.1478324055671692,
      "num_tokens": 16635068.0,
      "step": 8000
    },
    {
      "entropy": 5.870395565032959,
      "epoch": 0.7261429608127722,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004954502078990623,
      "loss": 5.6539,
      "mean_token_accuracy": 0.15322620123624803,
      "num_tokens": 16645021.0,
      "step": 8005
    },
    {
      "entropy": 5.84760684967041,
      "epoch": 0.7265965166908563,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004954437316105457,
      "loss": 5.7219,
      "mean_token_accuracy": 0.14631387591362,
      "num_tokens": 16656021.0,
      "step": 8010
    },
    {
      "entropy": 5.818306398391724,
      "epoch": 0.7270500725689405,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004954372507631965,
      "loss": 5.6209,
      "mean_token_accuracy": 0.15864094197750092,
      "num_tokens": 16665236.0,
      "step": 8015
    },
    {
      "entropy": 5.840885925292969,
      "epoch": 0.7275036284470247,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004954307653571489,
      "loss": 5.8054,
      "mean_token_accuracy": 0.14249199628829956,
      "num_tokens": 16676690.0,
      "step": 8020
    },
    {
      "entropy": 5.868116760253907,
      "epoch": 0.7279571843251088,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004954242753925367,
      "loss": 5.6937,
      "mean_token_accuracy": 0.14094067886471748,
      "num_tokens": 16687288.0,
      "step": 8025
    },
    {
      "entropy": 5.850828790664673,
      "epoch": 0.7284107402031931,
      "grad_norm": 2.875,
      "learning_rate": 0.0004954177808694946,
      "loss": 5.7177,
      "mean_token_accuracy": 0.14539122730493545,
      "num_tokens": 16697873.0,
      "step": 8030
    },
    {
      "entropy": 5.858789253234863,
      "epoch": 0.7288642960812772,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004954112817881565,
      "loss": 5.7026,
      "mean_token_accuracy": 0.15312349796295166,
      "num_tokens": 16709002.0,
      "step": 8035
    },
    {
      "entropy": 5.85047926902771,
      "epoch": 0.7293178519593614,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004954047781486568,
      "loss": 5.6745,
      "mean_token_accuracy": 0.1481028601527214,
      "num_tokens": 16718989.0,
      "step": 8040
    },
    {
      "entropy": 5.945074272155762,
      "epoch": 0.7297714078374455,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004953982699511303,
      "loss": 5.7757,
      "mean_token_accuracy": 0.14789580926299095,
      "num_tokens": 16730425.0,
      "step": 8045
    },
    {
      "entropy": 5.838289260864258,
      "epoch": 0.7302249637155298,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004953917571957114,
      "loss": 5.6377,
      "mean_token_accuracy": 0.15277301967144014,
      "num_tokens": 16740216.0,
      "step": 8050
    },
    {
      "entropy": 5.768487119674683,
      "epoch": 0.730678519593614,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004953852398825346,
      "loss": 5.6622,
      "mean_token_accuracy": 0.15074789077043532,
      "num_tokens": 16752436.0,
      "step": 8055
    },
    {
      "entropy": 5.748122119903565,
      "epoch": 0.7311320754716981,
      "grad_norm": 3.453125,
      "learning_rate": 0.000495378718011735,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15310948193073273,
      "num_tokens": 16762631.0,
      "step": 8060
    },
    {
      "entropy": 5.8812836647033695,
      "epoch": 0.7315856313497823,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004953721915834473,
      "loss": 5.6898,
      "mean_token_accuracy": 0.1502485454082489,
      "num_tokens": 16772613.0,
      "step": 8065
    },
    {
      "entropy": 5.8909612655639645,
      "epoch": 0.7320391872278664,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004953656605978067,
      "loss": 5.7026,
      "mean_token_accuracy": 0.14999075829982758,
      "num_tokens": 16782963.0,
      "step": 8070
    },
    {
      "entropy": 5.819808101654052,
      "epoch": 0.7324927431059507,
      "grad_norm": 3.4375,
      "learning_rate": 0.000495359125054948,
      "loss": 5.6277,
      "mean_token_accuracy": 0.15619223564863205,
      "num_tokens": 16793892.0,
      "step": 8075
    },
    {
      "entropy": 5.7889190196990965,
      "epoch": 0.7329462989840348,
      "grad_norm": 2.75,
      "learning_rate": 0.0004953525849550063,
      "loss": 5.5992,
      "mean_token_accuracy": 0.15081862434744836,
      "num_tokens": 16803388.0,
      "step": 8080
    },
    {
      "entropy": 5.842611455917359,
      "epoch": 0.733399854862119,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004953460402981172,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1550586625933647,
      "num_tokens": 16813201.0,
      "step": 8085
    },
    {
      "entropy": 5.900736331939697,
      "epoch": 0.7338534107402032,
      "grad_norm": 2.125,
      "learning_rate": 0.0004953394910844157,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15295570343732834,
      "num_tokens": 16823265.0,
      "step": 8090
    },
    {
      "entropy": 5.833998298645019,
      "epoch": 0.7343069666182874,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004953329373140374,
      "loss": 5.5994,
      "mean_token_accuracy": 0.15627037435770036,
      "num_tokens": 16833057.0,
      "step": 8095
    },
    {
      "entropy": 5.8118898391723635,
      "epoch": 0.7347605224963716,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004953263789871179,
      "loss": 5.5855,
      "mean_token_accuracy": 0.15333315581083298,
      "num_tokens": 16843488.0,
      "step": 8100
    },
    {
      "entropy": 5.778717279434204,
      "epoch": 0.7352140783744557,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004953198161037925,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15874518752098082,
      "num_tokens": 16852448.0,
      "step": 8105
    },
    {
      "entropy": 5.782966566085816,
      "epoch": 0.7356676342525399,
      "grad_norm": 9.5625,
      "learning_rate": 0.0004953132486641975,
      "loss": 5.6685,
      "mean_token_accuracy": 0.15144542902708052,
      "num_tokens": 16863944.0,
      "step": 8110
    },
    {
      "entropy": 5.844910621643066,
      "epoch": 0.736121190130624,
      "grad_norm": 2.515625,
      "learning_rate": 0.000495306676668468,
      "loss": 5.5967,
      "mean_token_accuracy": 0.15626500099897384,
      "num_tokens": 16873923.0,
      "step": 8115
    },
    {
      "entropy": 5.809791517257691,
      "epoch": 0.7365747460087083,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004953001001167406,
      "loss": 5.6782,
      "mean_token_accuracy": 0.15416514575481416,
      "num_tokens": 16884373.0,
      "step": 8120
    },
    {
      "entropy": 5.8766303062438965,
      "epoch": 0.7370283018867925,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004952935190091508,
      "loss": 5.7971,
      "mean_token_accuracy": 0.14941829442977905,
      "num_tokens": 16895869.0,
      "step": 8125
    },
    {
      "entropy": 5.778482818603516,
      "epoch": 0.7374818577648766,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004952869333458349,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1520070217549801,
      "num_tokens": 16906737.0,
      "step": 8130
    },
    {
      "entropy": 5.758775615692139,
      "epoch": 0.7379354136429608,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004952803431269292,
      "loss": 5.5958,
      "mean_token_accuracy": 0.15876534283161164,
      "num_tokens": 16916083.0,
      "step": 8135
    },
    {
      "entropy": 5.893442249298095,
      "epoch": 0.738388969521045,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004952737483525699,
      "loss": 5.7493,
      "mean_token_accuracy": 0.14965495616197586,
      "num_tokens": 16926419.0,
      "step": 8140
    },
    {
      "entropy": 5.902286958694458,
      "epoch": 0.7388425253991292,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004952671490228932,
      "loss": 5.6993,
      "mean_token_accuracy": 0.14944136291742324,
      "num_tokens": 16936067.0,
      "step": 8145
    },
    {
      "entropy": 5.920256948471069,
      "epoch": 0.7392960812772134,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004952605451380357,
      "loss": 5.6441,
      "mean_token_accuracy": 0.1541039377450943,
      "num_tokens": 16946571.0,
      "step": 8150
    },
    {
      "entropy": 5.820447397232056,
      "epoch": 0.7397496371552975,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004952539366981342,
      "loss": 5.7088,
      "mean_token_accuracy": 0.14866316616535186,
      "num_tokens": 16956764.0,
      "step": 8155
    },
    {
      "entropy": 5.835457468032837,
      "epoch": 0.7402031930333817,
      "grad_norm": 2.375,
      "learning_rate": 0.000495247323703325,
      "loss": 5.6093,
      "mean_token_accuracy": 0.1589105397462845,
      "num_tokens": 16966002.0,
      "step": 8160
    },
    {
      "entropy": 5.838615369796753,
      "epoch": 0.7406567489114659,
      "grad_norm": 2.4375,
      "learning_rate": 0.000495240706153745,
      "loss": 5.7271,
      "mean_token_accuracy": 0.1552692323923111,
      "num_tokens": 16975705.0,
      "step": 8165
    },
    {
      "entropy": 5.8247520446777346,
      "epoch": 0.7411103047895501,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004952340840495312,
      "loss": 5.6383,
      "mean_token_accuracy": 0.15176112204790115,
      "num_tokens": 16985945.0,
      "step": 8170
    },
    {
      "entropy": 5.8564215183258055,
      "epoch": 0.7415638606676342,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004952274573908203,
      "loss": 5.6046,
      "mean_token_accuracy": 0.15430220663547517,
      "num_tokens": 16995475.0,
      "step": 8175
    },
    {
      "entropy": 5.7460408210754395,
      "epoch": 0.7420174165457184,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004952208261777496,
      "loss": 5.6529,
      "mean_token_accuracy": 0.15099769681692124,
      "num_tokens": 17005028.0,
      "step": 8180
    },
    {
      "entropy": 5.812251472473145,
      "epoch": 0.7424709724238027,
      "grad_norm": 2.40625,
      "learning_rate": 0.000495214190410456,
      "loss": 5.6912,
      "mean_token_accuracy": 0.15830041766166686,
      "num_tokens": 17014786.0,
      "step": 8185
    },
    {
      "entropy": 5.9062474250793455,
      "epoch": 0.7429245283018868,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004952075500890768,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1491142690181732,
      "num_tokens": 17024431.0,
      "step": 8190
    },
    {
      "entropy": 5.860283613204956,
      "epoch": 0.743378084179971,
      "grad_norm": 3.25,
      "learning_rate": 0.0004952009052137493,
      "loss": 5.6797,
      "mean_token_accuracy": 0.1560024857521057,
      "num_tokens": 17034368.0,
      "step": 8195
    },
    {
      "entropy": 5.803822708129883,
      "epoch": 0.7438316400580551,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004951942557846112,
      "loss": 5.7083,
      "mean_token_accuracy": 0.15151755660772323,
      "num_tokens": 17043818.0,
      "step": 8200
    },
    {
      "entropy": 5.9088828563690186,
      "epoch": 0.7442851959361393,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004951876018017997,
      "loss": 5.7013,
      "mean_token_accuracy": 0.15114538073539735,
      "num_tokens": 17054843.0,
      "step": 8205
    },
    {
      "entropy": 5.918909502029419,
      "epoch": 0.7447387518142236,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004951809432654524,
      "loss": 5.7167,
      "mean_token_accuracy": 0.15287568420171738,
      "num_tokens": 17065539.0,
      "step": 8210
    },
    {
      "entropy": 5.796352529525757,
      "epoch": 0.7451923076923077,
      "grad_norm": 2.84375,
      "learning_rate": 0.000495174280175707,
      "loss": 5.6057,
      "mean_token_accuracy": 0.1571377620100975,
      "num_tokens": 17075443.0,
      "step": 8215
    },
    {
      "entropy": 5.875583744049072,
      "epoch": 0.7456458635703919,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004951676125327015,
      "loss": 5.7409,
      "mean_token_accuracy": 0.15380513221025466,
      "num_tokens": 17085490.0,
      "step": 8220
    },
    {
      "entropy": 5.88790078163147,
      "epoch": 0.746099419448476,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004951609403365737,
      "loss": 5.7283,
      "mean_token_accuracy": 0.15716300457715987,
      "num_tokens": 17094803.0,
      "step": 8225
    },
    {
      "entropy": 5.826867723464966,
      "epoch": 0.7465529753265603,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004951542635874615,
      "loss": 5.5683,
      "mean_token_accuracy": 0.15698309987783432,
      "num_tokens": 17104373.0,
      "step": 8230
    },
    {
      "entropy": 5.838105249404907,
      "epoch": 0.7470065312046444,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004951475822855029,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1605461299419403,
      "num_tokens": 17115469.0,
      "step": 8235
    },
    {
      "entropy": 5.7935174942016605,
      "epoch": 0.7474600870827286,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004951408964308364,
      "loss": 5.6534,
      "mean_token_accuracy": 0.15251931995153428,
      "num_tokens": 17125443.0,
      "step": 8240
    },
    {
      "entropy": 5.811115503311157,
      "epoch": 0.7479136429608128,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004951342060236,
      "loss": 5.6107,
      "mean_token_accuracy": 0.15216806083917617,
      "num_tokens": 17136253.0,
      "step": 8245
    },
    {
      "entropy": 5.925781488418579,
      "epoch": 0.7483671988388969,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004951275110639321,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1573117569088936,
      "num_tokens": 17145544.0,
      "step": 8250
    },
    {
      "entropy": 5.821874618530273,
      "epoch": 0.7488207547169812,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004951208115519712,
      "loss": 5.6796,
      "mean_token_accuracy": 0.15335864275693895,
      "num_tokens": 17155043.0,
      "step": 8255
    },
    {
      "entropy": 5.746136140823364,
      "epoch": 0.7492743105950653,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004951141074878558,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1515687733888626,
      "num_tokens": 17164885.0,
      "step": 8260
    },
    {
      "entropy": 5.913036060333252,
      "epoch": 0.7497278664731495,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004951073988717246,
      "loss": 5.7102,
      "mean_token_accuracy": 0.15280442237854003,
      "num_tokens": 17175681.0,
      "step": 8265
    },
    {
      "entropy": 5.837083959579468,
      "epoch": 0.7501814223512336,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004951006857037163,
      "loss": 5.577,
      "mean_token_accuracy": 0.15716822296380997,
      "num_tokens": 17186002.0,
      "step": 8270
    },
    {
      "entropy": 5.79591007232666,
      "epoch": 0.7506349782293179,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004950939679839696,
      "loss": 5.634,
      "mean_token_accuracy": 0.1542076975107193,
      "num_tokens": 17195499.0,
      "step": 8275
    },
    {
      "entropy": 5.812315368652344,
      "epoch": 0.7510885341074021,
      "grad_norm": 9.1875,
      "learning_rate": 0.0004950872457126237,
      "loss": 5.6553,
      "mean_token_accuracy": 0.1538190796971321,
      "num_tokens": 17204684.0,
      "step": 8280
    },
    {
      "entropy": 5.837529897689819,
      "epoch": 0.7515420899854862,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004950805188898175,
      "loss": 5.6769,
      "mean_token_accuracy": 0.1444774940609932,
      "num_tokens": 17213987.0,
      "step": 8285
    },
    {
      "entropy": 5.825541210174561,
      "epoch": 0.7519956458635704,
      "grad_norm": 3.25,
      "learning_rate": 0.00049507378751569,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1459720402956009,
      "num_tokens": 17224512.0,
      "step": 8290
    },
    {
      "entropy": 5.912886762619019,
      "epoch": 0.7524492017416545,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004950670515903805,
      "loss": 5.7117,
      "mean_token_accuracy": 0.1544044181704521,
      "num_tokens": 17234362.0,
      "step": 8295
    },
    {
      "entropy": 5.879697990417481,
      "epoch": 0.7529027576197388,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004950603111140283,
      "loss": 5.7077,
      "mean_token_accuracy": 0.1570012852549553,
      "num_tokens": 17244241.0,
      "step": 8300
    },
    {
      "entropy": 5.806952905654907,
      "epoch": 0.753356313497823,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004950535660867728,
      "loss": 5.6538,
      "mean_token_accuracy": 0.15973338037729262,
      "num_tokens": 17254689.0,
      "step": 8305
    },
    {
      "entropy": 5.83306655883789,
      "epoch": 0.7538098693759071,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004950468165087535,
      "loss": 5.7449,
      "mean_token_accuracy": 0.14953402578830718,
      "num_tokens": 17265506.0,
      "step": 8310
    },
    {
      "entropy": 5.895908164978027,
      "epoch": 0.7542634252539913,
      "grad_norm": 5.375,
      "learning_rate": 0.0004950400623801098,
      "loss": 5.7896,
      "mean_token_accuracy": 0.1489114761352539,
      "num_tokens": 17276376.0,
      "step": 8315
    },
    {
      "entropy": 5.948290395736694,
      "epoch": 0.7547169811320755,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004950333037009818,
      "loss": 5.7097,
      "mean_token_accuracy": 0.14384376108646393,
      "num_tokens": 17287005.0,
      "step": 8320
    },
    {
      "entropy": 5.916612863540649,
      "epoch": 0.7551705370101597,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004950265404715089,
      "loss": 5.713,
      "mean_token_accuracy": 0.14719957262277603,
      "num_tokens": 17297595.0,
      "step": 8325
    },
    {
      "entropy": 5.76622200012207,
      "epoch": 0.7556240928882438,
      "grad_norm": 2.359375,
      "learning_rate": 0.000495019772691831,
      "loss": 5.5757,
      "mean_token_accuracy": 0.15127571076154708,
      "num_tokens": 17307319.0,
      "step": 8330
    },
    {
      "entropy": 5.819165468215942,
      "epoch": 0.756077648766328,
      "grad_norm": 8.3125,
      "learning_rate": 0.0004950130003620882,
      "loss": 5.7815,
      "mean_token_accuracy": 0.15138889104127884,
      "num_tokens": 17317978.0,
      "step": 8335
    },
    {
      "entropy": 5.835421848297119,
      "epoch": 0.7565312046444121,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004950062234824204,
      "loss": 5.6732,
      "mean_token_accuracy": 0.14981579184532165,
      "num_tokens": 17328339.0,
      "step": 8340
    },
    {
      "entropy": 5.840034532546997,
      "epoch": 0.7569847605224964,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004949994420529679,
      "loss": 5.7422,
      "mean_token_accuracy": 0.14759217202663422,
      "num_tokens": 17338668.0,
      "step": 8345
    },
    {
      "entropy": 5.747548532485962,
      "epoch": 0.7574383164005806,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004949926560738709,
      "loss": 5.6809,
      "mean_token_accuracy": 0.1471247635781765,
      "num_tokens": 17348609.0,
      "step": 8350
    },
    {
      "entropy": 5.8837117671966555,
      "epoch": 0.7578918722786647,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004949858655452697,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1528981566429138,
      "num_tokens": 17358014.0,
      "step": 8355
    },
    {
      "entropy": 5.897519922256469,
      "epoch": 0.7583454281567489,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004949790704673048,
      "loss": 5.7057,
      "mean_token_accuracy": 0.150955131649971,
      "num_tokens": 17367975.0,
      "step": 8360
    },
    {
      "entropy": 5.8025383949279785,
      "epoch": 0.7587989840348331,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004949722708401166,
      "loss": 5.6525,
      "mean_token_accuracy": 0.15427399575710296,
      "num_tokens": 17377960.0,
      "step": 8365
    },
    {
      "entropy": 5.839379596710205,
      "epoch": 0.7592525399129173,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004949654666638457,
      "loss": 5.7023,
      "mean_token_accuracy": 0.15956505686044692,
      "num_tokens": 17387334.0,
      "step": 8370
    },
    {
      "entropy": 5.810500812530518,
      "epoch": 0.7597060957910015,
      "grad_norm": 3.359375,
      "learning_rate": 0.000494958657938633,
      "loss": 5.6878,
      "mean_token_accuracy": 0.15003183782100676,
      "num_tokens": 17396521.0,
      "step": 8375
    },
    {
      "entropy": 5.75392165184021,
      "epoch": 0.7601596516690856,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004949518446646192,
      "loss": 5.6549,
      "mean_token_accuracy": 0.15234262943267823,
      "num_tokens": 17407217.0,
      "step": 8380
    },
    {
      "entropy": 5.897473382949829,
      "epoch": 0.7606132075471698,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004949450268419451,
      "loss": 5.7301,
      "mean_token_accuracy": 0.15099592208862306,
      "num_tokens": 17416982.0,
      "step": 8385
    },
    {
      "entropy": 5.7785039901733395,
      "epoch": 0.761066763425254,
      "grad_norm": 2.84375,
      "learning_rate": 0.000494938204470752,
      "loss": 5.6164,
      "mean_token_accuracy": 0.153751140832901,
      "num_tokens": 17427636.0,
      "step": 8390
    },
    {
      "entropy": 5.792123603820801,
      "epoch": 0.7615203193033382,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004949313775511806,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1500251740217209,
      "num_tokens": 17437488.0,
      "step": 8395
    },
    {
      "entropy": 5.826402997970581,
      "epoch": 0.7619738751814223,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004949245460833723,
      "loss": 5.5337,
      "mean_token_accuracy": 0.15660372674465178,
      "num_tokens": 17447838.0,
      "step": 8400
    },
    {
      "entropy": 5.9221526145935055,
      "epoch": 0.7624274310595065,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004949177100674684,
      "loss": 5.7284,
      "mean_token_accuracy": 0.15142788887023925,
      "num_tokens": 17457257.0,
      "step": 8405
    },
    {
      "entropy": 5.868713808059693,
      "epoch": 0.7628809869375908,
      "grad_norm": 3.75,
      "learning_rate": 0.0004949108695036102,
      "loss": 5.6654,
      "mean_token_accuracy": 0.1565401151776314,
      "num_tokens": 17467729.0,
      "step": 8410
    },
    {
      "entropy": 5.8393584251403805,
      "epoch": 0.7633345428156749,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004949040243919392,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1448308013379574,
      "num_tokens": 17477692.0,
      "step": 8415
    },
    {
      "entropy": 5.799105072021485,
      "epoch": 0.7637880986937591,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004948971747325968,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15476165413856507,
      "num_tokens": 17488730.0,
      "step": 8420
    },
    {
      "entropy": 5.824799346923828,
      "epoch": 0.7642416545718432,
      "grad_norm": 3.640625,
      "learning_rate": 0.000494890320525725,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1491442933678627,
      "num_tokens": 17498374.0,
      "step": 8425
    },
    {
      "entropy": 5.872576713562012,
      "epoch": 0.7646952104499274,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004948834617714652,
      "loss": 5.6221,
      "mean_token_accuracy": 0.16017669290304185,
      "num_tokens": 17508601.0,
      "step": 8430
    },
    {
      "entropy": 5.882133340835571,
      "epoch": 0.7651487663280117,
      "grad_norm": 3.875,
      "learning_rate": 0.0004948765984699595,
      "loss": 5.6619,
      "mean_token_accuracy": 0.15254097878932954,
      "num_tokens": 17518164.0,
      "step": 8435
    },
    {
      "entropy": 5.8255561828613285,
      "epoch": 0.7656023222060958,
      "grad_norm": 8.625,
      "learning_rate": 0.0004948697306213494,
      "loss": 5.6263,
      "mean_token_accuracy": 0.15994967967271806,
      "num_tokens": 17527244.0,
      "step": 8440
    },
    {
      "entropy": 5.820489692687988,
      "epoch": 0.76605587808418,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004948628582257775,
      "loss": 5.7001,
      "mean_token_accuracy": 0.15338760912418364,
      "num_tokens": 17538045.0,
      "step": 8445
    },
    {
      "entropy": 5.837648820877075,
      "epoch": 0.7665094339622641,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004948559812833857,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1533903732895851,
      "num_tokens": 17547584.0,
      "step": 8450
    },
    {
      "entropy": 5.848302984237671,
      "epoch": 0.7669629898403484,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004948490997943162,
      "loss": 5.6926,
      "mean_token_accuracy": 0.14964775890111923,
      "num_tokens": 17558925.0,
      "step": 8455
    },
    {
      "entropy": 5.825404071807862,
      "epoch": 0.7674165457184325,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004948422137587111,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1478324979543686,
      "num_tokens": 17569355.0,
      "step": 8460
    },
    {
      "entropy": 5.838606214523315,
      "epoch": 0.7678701015965167,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004948353231767131,
      "loss": 5.65,
      "mean_token_accuracy": 0.15104758143424987,
      "num_tokens": 17579782.0,
      "step": 8465
    },
    {
      "entropy": 5.851181888580323,
      "epoch": 0.7683236574746009,
      "grad_norm": 2.125,
      "learning_rate": 0.0004948284280484645,
      "loss": 5.6885,
      "mean_token_accuracy": 0.15970910787582399,
      "num_tokens": 17589772.0,
      "step": 8470
    },
    {
      "entropy": 5.735399580001831,
      "epoch": 0.768777213352685,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004948215283741081,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15218766033649445,
      "num_tokens": 17599613.0,
      "step": 8475
    },
    {
      "entropy": 5.824175262451172,
      "epoch": 0.7692307692307693,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004948146241537866,
      "loss": 5.704,
      "mean_token_accuracy": 0.15038759112358094,
      "num_tokens": 17609056.0,
      "step": 8480
    },
    {
      "entropy": 5.834938859939575,
      "epoch": 0.7696843251088534,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004948077153876425,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15458367764949799,
      "num_tokens": 17618620.0,
      "step": 8485
    },
    {
      "entropy": 5.845889282226563,
      "epoch": 0.7701378809869376,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004948008020758188,
      "loss": 5.5825,
      "mean_token_accuracy": 0.1617045596241951,
      "num_tokens": 17627630.0,
      "step": 8490
    },
    {
      "entropy": 5.747945404052734,
      "epoch": 0.7705914368650217,
      "grad_norm": 4.0,
      "learning_rate": 0.0004947938842184585,
      "loss": 5.615,
      "mean_token_accuracy": 0.15477360188961028,
      "num_tokens": 17638903.0,
      "step": 8495
    },
    {
      "entropy": 5.8060832023620605,
      "epoch": 0.771044992743106,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004947869618157048,
      "loss": 5.6517,
      "mean_token_accuracy": 0.15006260126829146,
      "num_tokens": 17649257.0,
      "step": 8500
    },
    {
      "entropy": 5.867640686035156,
      "epoch": 0.7714985486211902,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004947800348677006,
      "loss": 5.6267,
      "mean_token_accuracy": 0.16068045347929,
      "num_tokens": 17659495.0,
      "step": 8505
    },
    {
      "entropy": 5.778565263748169,
      "epoch": 0.7719521044992743,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004947731033745893,
      "loss": 5.607,
      "mean_token_accuracy": 0.15641812831163407,
      "num_tokens": 17669671.0,
      "step": 8510
    },
    {
      "entropy": 5.852364349365234,
      "epoch": 0.7724056603773585,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004947661673365142,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1515066832304001,
      "num_tokens": 17680927.0,
      "step": 8515
    },
    {
      "entropy": 5.812708568572998,
      "epoch": 0.7728592162554426,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004947592267536188,
      "loss": 5.6459,
      "mean_token_accuracy": 0.1577283501625061,
      "num_tokens": 17691378.0,
      "step": 8520
    },
    {
      "entropy": 5.806360816955566,
      "epoch": 0.7733127721335269,
      "grad_norm": 4.125,
      "learning_rate": 0.0004947522816260466,
      "loss": 5.7186,
      "mean_token_accuracy": 0.14888853952288628,
      "num_tokens": 17700942.0,
      "step": 8525
    },
    {
      "entropy": 5.797103643417358,
      "epoch": 0.773766328011611,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004947453319539411,
      "loss": 5.6122,
      "mean_token_accuracy": 0.15577647536993028,
      "num_tokens": 17710076.0,
      "step": 8530
    },
    {
      "entropy": 5.8159160137176515,
      "epoch": 0.7742198838896952,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004947383777374462,
      "loss": 5.6645,
      "mean_token_accuracy": 0.15285882651805877,
      "num_tokens": 17721009.0,
      "step": 8535
    },
    {
      "entropy": 5.78311185836792,
      "epoch": 0.7746734397677794,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004947314189767058,
      "loss": 5.6808,
      "mean_token_accuracy": 0.15498824268579484,
      "num_tokens": 17731821.0,
      "step": 8540
    },
    {
      "entropy": 5.730218362808228,
      "epoch": 0.7751269956458636,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004947244556718635,
      "loss": 5.56,
      "mean_token_accuracy": 0.16283880770206452,
      "num_tokens": 17741425.0,
      "step": 8545
    },
    {
      "entropy": 5.806784772872925,
      "epoch": 0.7755805515239478,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004947174878230635,
      "loss": 5.6207,
      "mean_token_accuracy": 0.15820681154727936,
      "num_tokens": 17752381.0,
      "step": 8550
    },
    {
      "entropy": 5.77905125617981,
      "epoch": 0.7760341074020319,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004947105154304498,
      "loss": 5.6129,
      "mean_token_accuracy": 0.15068303048610687,
      "num_tokens": 17762768.0,
      "step": 8555
    },
    {
      "entropy": 5.791327667236328,
      "epoch": 0.7764876632801161,
      "grad_norm": 3.125,
      "learning_rate": 0.0004947035384941667,
      "loss": 5.6467,
      "mean_token_accuracy": 0.15347497612237931,
      "num_tokens": 17774032.0,
      "step": 8560
    },
    {
      "entropy": 5.882103633880615,
      "epoch": 0.7769412191582002,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004946965570143584,
      "loss": 5.7596,
      "mean_token_accuracy": 0.14765684455633163,
      "num_tokens": 17785098.0,
      "step": 8565
    },
    {
      "entropy": 5.897441005706787,
      "epoch": 0.7773947750362845,
      "grad_norm": 2.625,
      "learning_rate": 0.0004946895709911695,
      "loss": 5.6493,
      "mean_token_accuracy": 0.15527293384075164,
      "num_tokens": 17795487.0,
      "step": 8570
    },
    {
      "entropy": 5.820744800567627,
      "epoch": 0.7778483309143687,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004946825804247442,
      "loss": 5.7713,
      "mean_token_accuracy": 0.14488444924354554,
      "num_tokens": 17806045.0,
      "step": 8575
    },
    {
      "entropy": 5.891991424560547,
      "epoch": 0.7783018867924528,
      "grad_norm": 3.296875,
      "learning_rate": 0.000494675585315227,
      "loss": 5.7503,
      "mean_token_accuracy": 0.14853301644325256,
      "num_tokens": 17816948.0,
      "step": 8580
    },
    {
      "entropy": 5.856547927856445,
      "epoch": 0.778755442670537,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004946685856627628,
      "loss": 5.6418,
      "mean_token_accuracy": 0.15747622698545455,
      "num_tokens": 17826929.0,
      "step": 8585
    },
    {
      "entropy": 5.876050043106079,
      "epoch": 0.7792089985486212,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004946615814674963,
      "loss": 5.7598,
      "mean_token_accuracy": 0.15428425073623658,
      "num_tokens": 17837780.0,
      "step": 8590
    },
    {
      "entropy": 5.781055736541748,
      "epoch": 0.7796625544267054,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004946545727295724,
      "loss": 5.533,
      "mean_token_accuracy": 0.15893158614635466,
      "num_tokens": 17847835.0,
      "step": 8595
    },
    {
      "entropy": 5.804688358306885,
      "epoch": 0.7801161103047896,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004946475594491359,
      "loss": 5.6283,
      "mean_token_accuracy": 0.15911484211683274,
      "num_tokens": 17858324.0,
      "step": 8600
    },
    {
      "entropy": 5.783213043212891,
      "epoch": 0.7805696661828737,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004946405416263319,
      "loss": 5.6679,
      "mean_token_accuracy": 0.1482814759016037,
      "num_tokens": 17868455.0,
      "step": 8605
    },
    {
      "entropy": 5.856879234313965,
      "epoch": 0.7810232220609579,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004946335192613054,
      "loss": 5.6486,
      "mean_token_accuracy": 0.15848930925130844,
      "num_tokens": 17879038.0,
      "step": 8610
    },
    {
      "entropy": 5.730532503128051,
      "epoch": 0.7814767779390421,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004946264923542019,
      "loss": 5.5259,
      "mean_token_accuracy": 0.15444619804620743,
      "num_tokens": 17888153.0,
      "step": 8615
    },
    {
      "entropy": 5.750776576995849,
      "epoch": 0.7819303338171263,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004946194609051664,
      "loss": 5.6526,
      "mean_token_accuracy": 0.15871244743466378,
      "num_tokens": 17898289.0,
      "step": 8620
    },
    {
      "entropy": 5.8846351146698,
      "epoch": 0.7823838896952104,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004946124249143447,
      "loss": 5.6435,
      "mean_token_accuracy": 0.16097084879875184,
      "num_tokens": 17909192.0,
      "step": 8625
    },
    {
      "entropy": 5.874616003036499,
      "epoch": 0.7828374455732946,
      "grad_norm": 3.40625,
      "learning_rate": 0.000494605384381882,
      "loss": 5.6155,
      "mean_token_accuracy": 0.15133094936609268,
      "num_tokens": 17918937.0,
      "step": 8630
    },
    {
      "entropy": 5.837487697601318,
      "epoch": 0.7832910014513788,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004945983393079239,
      "loss": 5.6594,
      "mean_token_accuracy": 0.16026734709739685,
      "num_tokens": 17929346.0,
      "step": 8635
    },
    {
      "entropy": 5.757245635986328,
      "epoch": 0.783744557329463,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004945912896926163,
      "loss": 5.6674,
      "mean_token_accuracy": 0.15303874313831328,
      "num_tokens": 17939479.0,
      "step": 8640
    },
    {
      "entropy": 5.875200843811035,
      "epoch": 0.7841981132075472,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004945842355361049,
      "loss": 5.741,
      "mean_token_accuracy": 0.14947696030139923,
      "num_tokens": 17949841.0,
      "step": 8645
    },
    {
      "entropy": 5.829590225219727,
      "epoch": 0.7846516690856313,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004945771768385355,
      "loss": 5.6726,
      "mean_token_accuracy": 0.15517295747995377,
      "num_tokens": 17960334.0,
      "step": 8650
    },
    {
      "entropy": 5.814268445968628,
      "epoch": 0.7851052249637155,
      "grad_norm": 4.59375,
      "learning_rate": 0.000494570113600054,
      "loss": 5.6043,
      "mean_token_accuracy": 0.15772475600242614,
      "num_tokens": 17970595.0,
      "step": 8655
    },
    {
      "entropy": 5.84750337600708,
      "epoch": 0.7855587808417998,
      "grad_norm": 6.25,
      "learning_rate": 0.0004945630458208068,
      "loss": 5.7539,
      "mean_token_accuracy": 0.14532526433467866,
      "num_tokens": 17981301.0,
      "step": 8660
    },
    {
      "entropy": 5.788166093826294,
      "epoch": 0.7860123367198839,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004945559735009397,
      "loss": 5.6026,
      "mean_token_accuracy": 0.1555872455239296,
      "num_tokens": 17991262.0,
      "step": 8665
    },
    {
      "entropy": 5.872292470932007,
      "epoch": 0.7864658925979681,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004945488966405993,
      "loss": 5.6259,
      "mean_token_accuracy": 0.15453413426876067,
      "num_tokens": 18001212.0,
      "step": 8670
    },
    {
      "entropy": 5.79869122505188,
      "epoch": 0.7869194484760522,
      "grad_norm": 7.03125,
      "learning_rate": 0.0004945418152399317,
      "loss": 5.6668,
      "mean_token_accuracy": 0.15686399787664412,
      "num_tokens": 18010432.0,
      "step": 8675
    },
    {
      "entropy": 5.775568866729737,
      "epoch": 0.7873730043541364,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004945347292990833,
      "loss": 5.6717,
      "mean_token_accuracy": 0.14896780550479888,
      "num_tokens": 18020902.0,
      "step": 8680
    },
    {
      "entropy": 5.856520271301269,
      "epoch": 0.7878265602322206,
      "grad_norm": 3.625,
      "learning_rate": 0.0004945276388182008,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1509336993098259,
      "num_tokens": 18031133.0,
      "step": 8685
    },
    {
      "entropy": 5.967869615554809,
      "epoch": 0.7882801161103048,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004945205437974309,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15307042598724366,
      "num_tokens": 18042050.0,
      "step": 8690
    },
    {
      "entropy": 5.793651819229126,
      "epoch": 0.788733671988389,
      "grad_norm": 3.75,
      "learning_rate": 0.0004945134442369201,
      "loss": 5.6197,
      "mean_token_accuracy": 0.1614017441868782,
      "num_tokens": 18051642.0,
      "step": 8695
    },
    {
      "entropy": 5.821029710769653,
      "epoch": 0.7891872278664731,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004945063401368154,
      "loss": 5.6847,
      "mean_token_accuracy": 0.14467733949422837,
      "num_tokens": 18062180.0,
      "step": 8700
    },
    {
      "entropy": 5.86651382446289,
      "epoch": 0.7896407837445574,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004944992314972637,
      "loss": 5.6323,
      "mean_token_accuracy": 0.15366172045469284,
      "num_tokens": 18072302.0,
      "step": 8705
    },
    {
      "entropy": 5.817298030853271,
      "epoch": 0.7900943396226415,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004944921183184118,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15914198458194734,
      "num_tokens": 18081824.0,
      "step": 8710
    },
    {
      "entropy": 5.774461650848389,
      "epoch": 0.7905478955007257,
      "grad_norm": 3.53125,
      "learning_rate": 0.000494485000600407,
      "loss": 5.6877,
      "mean_token_accuracy": 0.1560596004128456,
      "num_tokens": 18092696.0,
      "step": 8715
    },
    {
      "entropy": 5.762605333328247,
      "epoch": 0.7910014513788098,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004944778783433965,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1632349118590355,
      "num_tokens": 18101790.0,
      "step": 8720
    },
    {
      "entropy": 5.901923751831054,
      "epoch": 0.791455007256894,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004944707515475276,
      "loss": 5.7263,
      "mean_token_accuracy": 0.1511840596795082,
      "num_tokens": 18111572.0,
      "step": 8725
    },
    {
      "entropy": 5.903787612915039,
      "epoch": 0.7919085631349783,
      "grad_norm": 3.125,
      "learning_rate": 0.0004944636202129475,
      "loss": 5.7068,
      "mean_token_accuracy": 0.16270414739847183,
      "num_tokens": 18120650.0,
      "step": 8730
    },
    {
      "entropy": 5.817595291137695,
      "epoch": 0.7923621190130624,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004944564843398039,
      "loss": 5.5961,
      "mean_token_accuracy": 0.16382812857627868,
      "num_tokens": 18131055.0,
      "step": 8735
    },
    {
      "entropy": 5.860651016235352,
      "epoch": 0.7928156748911466,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004944493439282443,
      "loss": 5.6884,
      "mean_token_accuracy": 0.1546909973025322,
      "num_tokens": 18141423.0,
      "step": 8740
    },
    {
      "entropy": 5.7539262771606445,
      "epoch": 0.7932692307692307,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004944421989784162,
      "loss": 5.6177,
      "mean_token_accuracy": 0.1582576185464859,
      "num_tokens": 18152015.0,
      "step": 8745
    },
    {
      "entropy": 5.820552015304566,
      "epoch": 0.793722786647315,
      "grad_norm": 3.5,
      "learning_rate": 0.0004944350494904676,
      "loss": 5.6973,
      "mean_token_accuracy": 0.15252969190478324,
      "num_tokens": 18161934.0,
      "step": 8750
    },
    {
      "entropy": 5.889200210571289,
      "epoch": 0.7941763425253991,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004944278954645463,
      "loss": 5.5921,
      "mean_token_accuracy": 0.15543517842888832,
      "num_tokens": 18172287.0,
      "step": 8755
    },
    {
      "entropy": 5.862867212295532,
      "epoch": 0.7946298984034833,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004944207369008002,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1599373534321785,
      "num_tokens": 18181015.0,
      "step": 8760
    },
    {
      "entropy": 5.802975034713745,
      "epoch": 0.7950834542815675,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004944135737993773,
      "loss": 5.7508,
      "mean_token_accuracy": 0.15021200925111772,
      "num_tokens": 18192382.0,
      "step": 8765
    },
    {
      "entropy": 5.775304937362671,
      "epoch": 0.7955370101596516,
      "grad_norm": 3.625,
      "learning_rate": 0.0004944064061604258,
      "loss": 5.627,
      "mean_token_accuracy": 0.16368629038333893,
      "num_tokens": 18202597.0,
      "step": 8770
    },
    {
      "entropy": 5.934780597686768,
      "epoch": 0.7959905660377359,
      "grad_norm": 2.125,
      "learning_rate": 0.0004943992339840939,
      "loss": 5.6766,
      "mean_token_accuracy": 0.1494644358754158,
      "num_tokens": 18213756.0,
      "step": 8775
    },
    {
      "entropy": 5.783631753921509,
      "epoch": 0.79644412191582,
      "grad_norm": 2.96875,
      "learning_rate": 0.00049439205727053,
      "loss": 5.6229,
      "mean_token_accuracy": 0.14898163676261902,
      "num_tokens": 18223669.0,
      "step": 8780
    },
    {
      "entropy": 5.778256940841675,
      "epoch": 0.7968976777939042,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004943848760198823,
      "loss": 5.6404,
      "mean_token_accuracy": 0.15757868587970733,
      "num_tokens": 18234884.0,
      "step": 8785
    },
    {
      "entropy": 5.814851236343384,
      "epoch": 0.7973512336719883,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004943776902322995,
      "loss": 5.5969,
      "mean_token_accuracy": 0.16033483147621155,
      "num_tokens": 18245155.0,
      "step": 8790
    },
    {
      "entropy": 5.731310796737671,
      "epoch": 0.7978047895500726,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004943704999079302,
      "loss": 5.5437,
      "mean_token_accuracy": 0.15940094143152236,
      "num_tokens": 18254686.0,
      "step": 8795
    },
    {
      "entropy": 5.79235520362854,
      "epoch": 0.7982583454281568,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004943633050469231,
      "loss": 5.5433,
      "mean_token_accuracy": 0.15957473516464232,
      "num_tokens": 18265473.0,
      "step": 8800
    },
    {
      "entropy": 5.91292028427124,
      "epoch": 0.7987119013062409,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004943561056494269,
      "loss": 5.7515,
      "mean_token_accuracy": 0.15094480961561202,
      "num_tokens": 18276024.0,
      "step": 8805
    },
    {
      "entropy": 5.838796234130859,
      "epoch": 0.7991654571843251,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004943489017155906,
      "loss": 5.6977,
      "mean_token_accuracy": 0.14398460388183593,
      "num_tokens": 18287261.0,
      "step": 8810
    },
    {
      "entropy": 5.832255887985229,
      "epoch": 0.7996190130624092,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004943416932455631,
      "loss": 5.7026,
      "mean_token_accuracy": 0.15213215947151185,
      "num_tokens": 18299374.0,
      "step": 8815
    },
    {
      "entropy": 5.7932600498199465,
      "epoch": 0.8000725689404935,
      "grad_norm": 2.25,
      "learning_rate": 0.0004943344802394935,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1507257863879204,
      "num_tokens": 18309344.0,
      "step": 8820
    },
    {
      "entropy": 5.860401153564453,
      "epoch": 0.8005261248185777,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004943272626975309,
      "loss": 5.6701,
      "mean_token_accuracy": 0.15301405191421508,
      "num_tokens": 18319778.0,
      "step": 8825
    },
    {
      "entropy": 5.827208948135376,
      "epoch": 0.8009796806966618,
      "grad_norm": 2.875,
      "learning_rate": 0.0004943200406198246,
      "loss": 5.6558,
      "mean_token_accuracy": 0.1550200790166855,
      "num_tokens": 18329787.0,
      "step": 8830
    },
    {
      "entropy": 5.767522525787354,
      "epoch": 0.801433236574746,
      "grad_norm": 2.609375,
      "learning_rate": 0.000494312814006524,
      "loss": 5.6147,
      "mean_token_accuracy": 0.16383739709854125,
      "num_tokens": 18340670.0,
      "step": 8835
    },
    {
      "entropy": 5.788567638397216,
      "epoch": 0.8018867924528302,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004943055828577784,
      "loss": 5.6304,
      "mean_token_accuracy": 0.14750366061925888,
      "num_tokens": 18352484.0,
      "step": 8840
    },
    {
      "entropy": 5.787442207336426,
      "epoch": 0.8023403483309144,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004942983471737376,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1558992326259613,
      "num_tokens": 18363103.0,
      "step": 8845
    },
    {
      "entropy": 5.772452688217163,
      "epoch": 0.8027939042089985,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004942911069545511,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1594387948513031,
      "num_tokens": 18373925.0,
      "step": 8850
    },
    {
      "entropy": 5.807936382293701,
      "epoch": 0.8032474600870827,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004942838622003686,
      "loss": 5.701,
      "mean_token_accuracy": 0.1509519785642624,
      "num_tokens": 18384081.0,
      "step": 8855
    },
    {
      "entropy": 5.827019786834716,
      "epoch": 0.8037010159651669,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004942766129113399,
      "loss": 5.5911,
      "mean_token_accuracy": 0.15893699526786803,
      "num_tokens": 18392510.0,
      "step": 8860
    },
    {
      "entropy": 5.900896501541138,
      "epoch": 0.8041545718432511,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004942693590876151,
      "loss": 5.7146,
      "mean_token_accuracy": 0.1524460308253765,
      "num_tokens": 18402809.0,
      "step": 8865
    },
    {
      "entropy": 5.886493778228759,
      "epoch": 0.8046081277213353,
      "grad_norm": 2.578125,
      "learning_rate": 0.000494262100729344,
      "loss": 5.7361,
      "mean_token_accuracy": 0.1542189508676529,
      "num_tokens": 18413342.0,
      "step": 8870
    },
    {
      "entropy": 5.874359846115112,
      "epoch": 0.8050616835994194,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004942548378366768,
      "loss": 5.7501,
      "mean_token_accuracy": 0.15199500173330308,
      "num_tokens": 18424735.0,
      "step": 8875
    },
    {
      "entropy": 5.829610395431518,
      "epoch": 0.8055152394775036,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004942475704097635,
      "loss": 5.6184,
      "mean_token_accuracy": 0.15787999778985978,
      "num_tokens": 18434797.0,
      "step": 8880
    },
    {
      "entropy": 5.796105766296387,
      "epoch": 0.8059687953555879,
      "grad_norm": 6.125,
      "learning_rate": 0.0004942402984487548,
      "loss": 5.6877,
      "mean_token_accuracy": 0.14930816888809204,
      "num_tokens": 18445456.0,
      "step": 8885
    },
    {
      "entropy": 5.800718307495117,
      "epoch": 0.806422351233672,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004942330219538008,
      "loss": 5.6542,
      "mean_token_accuracy": 0.14945662915706634,
      "num_tokens": 18456728.0,
      "step": 8890
    },
    {
      "entropy": 6.033436584472656,
      "epoch": 0.8068759071117562,
      "grad_norm": 2.78125,
      "learning_rate": 0.000494225740925052,
      "loss": 5.7818,
      "mean_token_accuracy": 0.14793658256530762,
      "num_tokens": 18468813.0,
      "step": 8895
    },
    {
      "entropy": 5.932030200958252,
      "epoch": 0.8073294629898403,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004942184553626591,
      "loss": 5.6778,
      "mean_token_accuracy": 0.154929731041193,
      "num_tokens": 18477912.0,
      "step": 8900
    },
    {
      "entropy": 5.841903495788574,
      "epoch": 0.8077830188679245,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004942111652667726,
      "loss": 5.6177,
      "mean_token_accuracy": 0.15868613123893738,
      "num_tokens": 18488115.0,
      "step": 8905
    },
    {
      "entropy": 5.836884880065918,
      "epoch": 0.8082365747460087,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004942038706375433,
      "loss": 5.6752,
      "mean_token_accuracy": 0.15271092504262923,
      "num_tokens": 18498853.0,
      "step": 8910
    },
    {
      "entropy": 5.817429161071777,
      "epoch": 0.8086901306240929,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004941965714751222,
      "loss": 5.7144,
      "mean_token_accuracy": 0.1488034538924694,
      "num_tokens": 18510047.0,
      "step": 8915
    },
    {
      "entropy": 5.796209287643433,
      "epoch": 0.809143686502177,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004941892677796601,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1562424249947071,
      "num_tokens": 18519673.0,
      "step": 8920
    },
    {
      "entropy": 5.834105730056763,
      "epoch": 0.8095972423802612,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004941819595513081,
      "loss": 5.7058,
      "mean_token_accuracy": 0.15310655534267426,
      "num_tokens": 18530090.0,
      "step": 8925
    },
    {
      "entropy": 5.897770929336548,
      "epoch": 0.8100507982583455,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004941746467902173,
      "loss": 5.7393,
      "mean_token_accuracy": 0.14204500168561934,
      "num_tokens": 18541862.0,
      "step": 8930
    },
    {
      "entropy": 5.893235349655152,
      "epoch": 0.8105043541364296,
      "grad_norm": 2.53125,
      "learning_rate": 0.000494167329496539,
      "loss": 5.713,
      "mean_token_accuracy": 0.14384791254997253,
      "num_tokens": 18551624.0,
      "step": 8935
    },
    {
      "entropy": 5.793308925628662,
      "epoch": 0.8109579100145138,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004941600076704244,
      "loss": 5.6614,
      "mean_token_accuracy": 0.15391065627336503,
      "num_tokens": 18561784.0,
      "step": 8940
    },
    {
      "entropy": 5.871464252471924,
      "epoch": 0.8114114658925979,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004941526813120251,
      "loss": 5.7296,
      "mean_token_accuracy": 0.15403200089931487,
      "num_tokens": 18572110.0,
      "step": 8945
    },
    {
      "entropy": 5.898204517364502,
      "epoch": 0.8118650217706821,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004941453504214924,
      "loss": 5.688,
      "mean_token_accuracy": 0.14993584677577018,
      "num_tokens": 18582012.0,
      "step": 8950
    },
    {
      "entropy": 5.844607734680176,
      "epoch": 0.8123185776487664,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004941380149989781,
      "loss": 5.5412,
      "mean_token_accuracy": 0.16359425187110901,
      "num_tokens": 18592850.0,
      "step": 8955
    },
    {
      "entropy": 5.773560285568237,
      "epoch": 0.8127721335268505,
      "grad_norm": 2.5,
      "learning_rate": 0.0004941306750446336,
      "loss": 5.6954,
      "mean_token_accuracy": 0.1542632609605789,
      "num_tokens": 18602590.0,
      "step": 8960
    },
    {
      "entropy": 5.790944147109985,
      "epoch": 0.8132256894049347,
      "grad_norm": 3.3125,
      "learning_rate": 0.000494123330558611,
      "loss": 5.5988,
      "mean_token_accuracy": 0.15769584104418755,
      "num_tokens": 18612607.0,
      "step": 8965
    },
    {
      "entropy": 5.793989849090576,
      "epoch": 0.8136792452830188,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004941159815410622,
      "loss": 5.5872,
      "mean_token_accuracy": 0.15930003374814988,
      "num_tokens": 18622936.0,
      "step": 8970
    },
    {
      "entropy": 5.842741632461548,
      "epoch": 0.8141328011611031,
      "grad_norm": 2.8125,
      "learning_rate": 0.000494108627992139,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1470210388302803,
      "num_tokens": 18633997.0,
      "step": 8975
    },
    {
      "entropy": 5.867417573928833,
      "epoch": 0.8145863570391872,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004941012699119936,
      "loss": 5.5763,
      "mean_token_accuracy": 0.1638900578022003,
      "num_tokens": 18644347.0,
      "step": 8980
    },
    {
      "entropy": 5.838050317764282,
      "epoch": 0.8150399129172714,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004940939073007779,
      "loss": 5.6593,
      "mean_token_accuracy": 0.15335242450237274,
      "num_tokens": 18654989.0,
      "step": 8985
    },
    {
      "entropy": 5.7650409698486325,
      "epoch": 0.8154934687953556,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004940865401586446,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15253210663795472,
      "num_tokens": 18665526.0,
      "step": 8990
    },
    {
      "entropy": 5.834532165527344,
      "epoch": 0.8159470246734397,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004940791684857458,
      "loss": 5.6554,
      "mean_token_accuracy": 0.15907703042030336,
      "num_tokens": 18676226.0,
      "step": 8995
    },
    {
      "entropy": 5.847899627685547,
      "epoch": 0.816400580551524,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004940717922822339,
      "loss": 5.5808,
      "mean_token_accuracy": 0.15463961511850358,
      "num_tokens": 18686499.0,
      "step": 9000
    },
    {
      "epoch": 0.816400580551524,
      "eval_entropy": 5.683191412606391,
      "eval_loss": 5.684431552886963,
      "eval_mean_token_accuracy": 0.15958393904131868,
      "eval_num_tokens": 18686499.0,
      "eval_runtime": 26.6876,
      "eval_samples_per_second": 1324.96,
      "eval_steps_per_second": 165.62,
      "step": 9000
    },
    {
      "entropy": 5.843375539779663,
      "epoch": 0.8168541364296081,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004940644115482615,
      "loss": 5.6417,
      "mean_token_accuracy": 0.15436051040887833,
      "num_tokens": 18696543.0,
      "step": 9005
    },
    {
      "entropy": 5.793101692199707,
      "epoch": 0.8173076923076923,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004940570262839812,
      "loss": 5.6896,
      "mean_token_accuracy": 0.15068877935409547,
      "num_tokens": 18707029.0,
      "step": 9010
    },
    {
      "entropy": 5.8139701843261715,
      "epoch": 0.8177612481857764,
      "grad_norm": 2.625,
      "learning_rate": 0.0004940496364895457,
      "loss": 5.6789,
      "mean_token_accuracy": 0.15451287627220153,
      "num_tokens": 18716937.0,
      "step": 9015
    },
    {
      "entropy": 5.897455024719238,
      "epoch": 0.8182148040638607,
      "grad_norm": 4.25,
      "learning_rate": 0.0004940422421651082,
      "loss": 5.6724,
      "mean_token_accuracy": 0.15182604044675826,
      "num_tokens": 18727478.0,
      "step": 9020
    },
    {
      "entropy": 5.864028549194336,
      "epoch": 0.8186683599419449,
      "grad_norm": 3.25,
      "learning_rate": 0.0004940348433108211,
      "loss": 5.5951,
      "mean_token_accuracy": 0.14794762730598449,
      "num_tokens": 18737339.0,
      "step": 9025
    },
    {
      "entropy": 5.851188659667969,
      "epoch": 0.819121915820029,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004940274399268377,
      "loss": 5.7162,
      "mean_token_accuracy": 0.15362059324979782,
      "num_tokens": 18747810.0,
      "step": 9030
    },
    {
      "entropy": 5.7702428817749025,
      "epoch": 0.8195754716981132,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004940200320133109,
      "loss": 5.6525,
      "mean_token_accuracy": 0.15094198137521744,
      "num_tokens": 18757564.0,
      "step": 9035
    },
    {
      "entropy": 5.884185075759888,
      "epoch": 0.8200290275761973,
      "grad_norm": 6.25,
      "learning_rate": 0.0004940126195703941,
      "loss": 5.7132,
      "mean_token_accuracy": 0.1497935563325882,
      "num_tokens": 18768196.0,
      "step": 9040
    },
    {
      "entropy": 5.898662471771241,
      "epoch": 0.8204825834542816,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004940052025982405,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16041729897260665,
      "num_tokens": 18778350.0,
      "step": 9045
    },
    {
      "entropy": 5.814507389068604,
      "epoch": 0.8209361393323658,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004939977810970035,
      "loss": 5.6194,
      "mean_token_accuracy": 0.14908331036567687,
      "num_tokens": 18789825.0,
      "step": 9050
    },
    {
      "entropy": 5.8118775367736815,
      "epoch": 0.8213896952104499,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004939903550668366,
      "loss": 5.6354,
      "mean_token_accuracy": 0.1541058212518692,
      "num_tokens": 18799793.0,
      "step": 9055
    },
    {
      "entropy": 5.806805038452149,
      "epoch": 0.8218432510885341,
      "grad_norm": 5.65625,
      "learning_rate": 0.0004939829245078934,
      "loss": 5.6747,
      "mean_token_accuracy": 0.15679603368043898,
      "num_tokens": 18809771.0,
      "step": 9060
    },
    {
      "entropy": 5.904952669143677,
      "epoch": 0.8222968069666183,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004939754894203275,
      "loss": 5.6922,
      "mean_token_accuracy": 0.1526167705655098,
      "num_tokens": 18820529.0,
      "step": 9065
    },
    {
      "entropy": 5.800691223144531,
      "epoch": 0.8227503628447025,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004939680498042926,
      "loss": 5.607,
      "mean_token_accuracy": 0.15987065434455872,
      "num_tokens": 18830797.0,
      "step": 9070
    },
    {
      "entropy": 5.785319471359253,
      "epoch": 0.8232039187227866,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004939606056599428,
      "loss": 5.6072,
      "mean_token_accuracy": 0.15448503717780113,
      "num_tokens": 18840526.0,
      "step": 9075
    },
    {
      "entropy": 5.772441053390503,
      "epoch": 0.8236574746008708,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004939531569874316,
      "loss": 5.6298,
      "mean_token_accuracy": 0.15375131517648696,
      "num_tokens": 18851201.0,
      "step": 9080
    },
    {
      "entropy": 5.755581855773926,
      "epoch": 0.824111030478955,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004939457037869135,
      "loss": 5.469,
      "mean_token_accuracy": 0.16889793276786805,
      "num_tokens": 18861676.0,
      "step": 9085
    },
    {
      "entropy": 5.7853155612945555,
      "epoch": 0.8245645863570392,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004939382460585424,
      "loss": 5.5963,
      "mean_token_accuracy": 0.15494900494813918,
      "num_tokens": 18872669.0,
      "step": 9090
    },
    {
      "entropy": 5.752126216888428,
      "epoch": 0.8250181422351234,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004939307838024726,
      "loss": 5.6265,
      "mean_token_accuracy": 0.15291281417012215,
      "num_tokens": 18882130.0,
      "step": 9095
    },
    {
      "entropy": 5.73763689994812,
      "epoch": 0.8254716981132075,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004939233170188585,
      "loss": 5.6226,
      "mean_token_accuracy": 0.15411534905433655,
      "num_tokens": 18892457.0,
      "step": 9100
    },
    {
      "entropy": 5.894270277023315,
      "epoch": 0.8259252539912917,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004939158457078543,
      "loss": 5.6544,
      "mean_token_accuracy": 0.15148564502596856,
      "num_tokens": 18903152.0,
      "step": 9105
    },
    {
      "entropy": 5.817868185043335,
      "epoch": 0.826378809869376,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004939083698696146,
      "loss": 5.6276,
      "mean_token_accuracy": 0.15088266208767892,
      "num_tokens": 18913740.0,
      "step": 9110
    },
    {
      "entropy": 5.8163670063018795,
      "epoch": 0.8268323657474601,
      "grad_norm": 3.9375,
      "learning_rate": 0.000493900889504294,
      "loss": 5.6636,
      "mean_token_accuracy": 0.15680994540452958,
      "num_tokens": 18923501.0,
      "step": 9115
    },
    {
      "entropy": 5.822394514083863,
      "epoch": 0.8272859216255443,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004938934046120473,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15569230169057846,
      "num_tokens": 18933255.0,
      "step": 9120
    },
    {
      "entropy": 5.795421028137207,
      "epoch": 0.8277394775036284,
      "grad_norm": 3.25,
      "learning_rate": 0.0004938859151930293,
      "loss": 5.6055,
      "mean_token_accuracy": 0.15849126726388932,
      "num_tokens": 18943649.0,
      "step": 9125
    },
    {
      "entropy": 5.788185548782349,
      "epoch": 0.8281930333817126,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004938784212473947,
      "loss": 5.5362,
      "mean_token_accuracy": 0.1615826740860939,
      "num_tokens": 18953187.0,
      "step": 9130
    },
    {
      "entropy": 5.836130285263062,
      "epoch": 0.8286465892597968,
      "grad_norm": 15.75,
      "learning_rate": 0.0004938709227752985,
      "loss": 5.6835,
      "mean_token_accuracy": 0.15411966443061828,
      "num_tokens": 18963015.0,
      "step": 9135
    },
    {
      "entropy": 5.836290073394776,
      "epoch": 0.829100145137881,
      "grad_norm": 3.21875,
      "learning_rate": 0.000493863419776896,
      "loss": 5.685,
      "mean_token_accuracy": 0.14958571940660476,
      "num_tokens": 18974184.0,
      "step": 9140
    },
    {
      "entropy": 5.796621084213257,
      "epoch": 0.8295537010159652,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004938559122523421,
      "loss": 5.6169,
      "mean_token_accuracy": 0.1585509866476059,
      "num_tokens": 18985036.0,
      "step": 9145
    },
    {
      "entropy": 5.808863782882691,
      "epoch": 0.8300072568940493,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004938484002017921,
      "loss": 5.7302,
      "mean_token_accuracy": 0.1534656725823879,
      "num_tokens": 18995237.0,
      "step": 9150
    },
    {
      "entropy": 5.818014335632324,
      "epoch": 0.8304608127721336,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004938408836254015,
      "loss": 5.73,
      "mean_token_accuracy": 0.14893487244844436,
      "num_tokens": 19006272.0,
      "step": 9155
    },
    {
      "entropy": 5.842798709869385,
      "epoch": 0.8309143686502177,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004938333625233258,
      "loss": 5.6512,
      "mean_token_accuracy": 0.1608937107026577,
      "num_tokens": 19016728.0,
      "step": 9160
    },
    {
      "entropy": 5.886384630203247,
      "epoch": 0.8313679245283019,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004938258368957202,
      "loss": 5.6925,
      "mean_token_accuracy": 0.14620290994644164,
      "num_tokens": 19028024.0,
      "step": 9165
    },
    {
      "entropy": 5.8492639541625975,
      "epoch": 0.831821480406386,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004938183067427406,
      "loss": 5.628,
      "mean_token_accuracy": 0.1614862248301506,
      "num_tokens": 19038218.0,
      "step": 9170
    },
    {
      "entropy": 5.802521514892578,
      "epoch": 0.8322750362844702,
      "grad_norm": 5.125,
      "learning_rate": 0.0004938107720645426,
      "loss": 5.675,
      "mean_token_accuracy": 0.15266878306865692,
      "num_tokens": 19048574.0,
      "step": 9175
    },
    {
      "entropy": 5.856499767303466,
      "epoch": 0.8327285921625545,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004938032328612821,
      "loss": 5.6296,
      "mean_token_accuracy": 0.15636837631464004,
      "num_tokens": 19057721.0,
      "step": 9180
    },
    {
      "entropy": 5.860321521759033,
      "epoch": 0.8331821480406386,
      "grad_norm": 3.59375,
      "learning_rate": 0.000493795689133115,
      "loss": 5.6015,
      "mean_token_accuracy": 0.1484677277505398,
      "num_tokens": 19067582.0,
      "step": 9185
    },
    {
      "entropy": 5.790124559402466,
      "epoch": 0.8336357039187228,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004937881408801974,
      "loss": 5.5827,
      "mean_token_accuracy": 0.15845344811677933,
      "num_tokens": 19077896.0,
      "step": 9190
    },
    {
      "entropy": 5.808318042755127,
      "epoch": 0.8340892597968069,
      "grad_norm": 2.640625,
      "learning_rate": 0.000493780588102685,
      "loss": 5.6812,
      "mean_token_accuracy": 0.15800550431013108,
      "num_tokens": 19088082.0,
      "step": 9195
    },
    {
      "entropy": 5.863030862808228,
      "epoch": 0.8345428156748912,
      "grad_norm": 3.0,
      "learning_rate": 0.0004937730308007346,
      "loss": 5.7038,
      "mean_token_accuracy": 0.151945798099041,
      "num_tokens": 19097176.0,
      "step": 9200
    },
    {
      "entropy": 5.761180305480957,
      "epoch": 0.8349963715529753,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004937654689745021,
      "loss": 5.6062,
      "mean_token_accuracy": 0.15972303003072738,
      "num_tokens": 19107835.0,
      "step": 9205
    },
    {
      "entropy": 5.709291219711304,
      "epoch": 0.8354499274310595,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004937579026241439,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1632474735379219,
      "num_tokens": 19118028.0,
      "step": 9210
    },
    {
      "entropy": 5.85373101234436,
      "epoch": 0.8359034833091437,
      "grad_norm": 13.3125,
      "learning_rate": 0.0004937503317498165,
      "loss": 5.6329,
      "mean_token_accuracy": 0.15670134127140045,
      "num_tokens": 19129223.0,
      "step": 9215
    },
    {
      "entropy": 5.874120140075684,
      "epoch": 0.8363570391872278,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004937427563516765,
      "loss": 5.7367,
      "mean_token_accuracy": 0.15082149356603622,
      "num_tokens": 19140048.0,
      "step": 9220
    },
    {
      "entropy": 5.816678190231324,
      "epoch": 0.8368105950653121,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004937351764298805,
      "loss": 5.7069,
      "mean_token_accuracy": 0.14715564996004105,
      "num_tokens": 19150853.0,
      "step": 9225
    },
    {
      "entropy": 5.82272515296936,
      "epoch": 0.8372641509433962,
      "grad_norm": 7.46875,
      "learning_rate": 0.0004937275919845855,
      "loss": 5.681,
      "mean_token_accuracy": 0.15349098891019822,
      "num_tokens": 19161645.0,
      "step": 9230
    },
    {
      "entropy": 5.871696996688843,
      "epoch": 0.8377177068214804,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004937200030159481,
      "loss": 5.7667,
      "mean_token_accuracy": 0.14946603924036025,
      "num_tokens": 19173276.0,
      "step": 9235
    },
    {
      "entropy": 5.915431976318359,
      "epoch": 0.8381712626995645,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004937124095241252,
      "loss": 5.6702,
      "mean_token_accuracy": 0.1593783065676689,
      "num_tokens": 19182949.0,
      "step": 9240
    },
    {
      "entropy": 5.884314680099488,
      "epoch": 0.8386248185776488,
      "grad_norm": 5.90625,
      "learning_rate": 0.0004937048115092739,
      "loss": 5.651,
      "mean_token_accuracy": 0.15436612367630004,
      "num_tokens": 19193504.0,
      "step": 9245
    },
    {
      "entropy": 5.835222673416138,
      "epoch": 0.839078374455733,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004936972089715515,
      "loss": 5.7284,
      "mean_token_accuracy": 0.14974120259284973,
      "num_tokens": 19203406.0,
      "step": 9250
    },
    {
      "entropy": 5.815620374679566,
      "epoch": 0.8395319303338171,
      "grad_norm": 4.03125,
      "learning_rate": 0.000493689601911115,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15398818552494048,
      "num_tokens": 19214325.0,
      "step": 9255
    },
    {
      "entropy": 5.845991325378418,
      "epoch": 0.8399854862119013,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004936819903281219,
      "loss": 5.7333,
      "mean_token_accuracy": 0.14285178184509278,
      "num_tokens": 19226466.0,
      "step": 9260
    },
    {
      "entropy": 5.934034585952759,
      "epoch": 0.8404390420899854,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004936743742227295,
      "loss": 5.7372,
      "mean_token_accuracy": 0.15123442113399505,
      "num_tokens": 19237032.0,
      "step": 9265
    },
    {
      "entropy": 5.916327333450317,
      "epoch": 0.8408925979680697,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004936667535950952,
      "loss": 5.7308,
      "mean_token_accuracy": 0.14760565385222435,
      "num_tokens": 19248009.0,
      "step": 9270
    },
    {
      "entropy": 5.850819635391235,
      "epoch": 0.8413461538461539,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004936591284453768,
      "loss": 5.7935,
      "mean_token_accuracy": 0.14887545853853226,
      "num_tokens": 19259363.0,
      "step": 9275
    },
    {
      "entropy": 5.817532968521118,
      "epoch": 0.841799709724238,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004936514987737319,
      "loss": 5.6921,
      "mean_token_accuracy": 0.15453299582004548,
      "num_tokens": 19269191.0,
      "step": 9280
    },
    {
      "entropy": 5.868520593643188,
      "epoch": 0.8422532656023222,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004936438645803182,
      "loss": 5.6331,
      "mean_token_accuracy": 0.15244528353214265,
      "num_tokens": 19280278.0,
      "step": 9285
    },
    {
      "entropy": 5.887366008758545,
      "epoch": 0.8427068214804064,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004936362258652938,
      "loss": 5.6228,
      "mean_token_accuracy": 0.15615448504686355,
      "num_tokens": 19290130.0,
      "step": 9290
    },
    {
      "entropy": 5.975782442092895,
      "epoch": 0.8431603773584906,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004936285826288165,
      "loss": 5.7635,
      "mean_token_accuracy": 0.15192786306142808,
      "num_tokens": 19301669.0,
      "step": 9295
    },
    {
      "entropy": 5.813149261474609,
      "epoch": 0.8436139332365747,
      "grad_norm": 3.0,
      "learning_rate": 0.0004936209348710444,
      "loss": 5.6288,
      "mean_token_accuracy": 0.15411185771226882,
      "num_tokens": 19312490.0,
      "step": 9300
    },
    {
      "entropy": 5.775185203552246,
      "epoch": 0.8440674891146589,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004936132825921355,
      "loss": 5.6643,
      "mean_token_accuracy": 0.1526573419570923,
      "num_tokens": 19321743.0,
      "step": 9305
    },
    {
      "entropy": 5.795629644393921,
      "epoch": 0.844521044992743,
      "grad_norm": 2.375,
      "learning_rate": 0.0004936056257922484,
      "loss": 5.5721,
      "mean_token_accuracy": 0.1604461818933487,
      "num_tokens": 19332149.0,
      "step": 9310
    },
    {
      "entropy": 5.8589396476745605,
      "epoch": 0.8449746008708273,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004935979644715412,
      "loss": 5.6545,
      "mean_token_accuracy": 0.15000626295804978,
      "num_tokens": 19341655.0,
      "step": 9315
    },
    {
      "entropy": 5.888137102127075,
      "epoch": 0.8454281567489115,
      "grad_norm": 3.0,
      "learning_rate": 0.0004935902986301724,
      "loss": 5.7102,
      "mean_token_accuracy": 0.149715293943882,
      "num_tokens": 19351404.0,
      "step": 9320
    },
    {
      "entropy": 5.82078709602356,
      "epoch": 0.8458817126269956,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004935826282683005,
      "loss": 5.7112,
      "mean_token_accuracy": 0.1484207406640053,
      "num_tokens": 19361788.0,
      "step": 9325
    },
    {
      "entropy": 5.812763786315918,
      "epoch": 0.8463352685050798,
      "grad_norm": 2.375,
      "learning_rate": 0.0004935749533860842,
      "loss": 5.6817,
      "mean_token_accuracy": 0.15633453130722047,
      "num_tokens": 19372902.0,
      "step": 9330
    },
    {
      "entropy": 5.739981412887573,
      "epoch": 0.846788824383164,
      "grad_norm": 2.3125,
      "learning_rate": 0.000493567273983682,
      "loss": 5.5744,
      "mean_token_accuracy": 0.1500100888311863,
      "num_tokens": 19382181.0,
      "step": 9335
    },
    {
      "entropy": 5.848348140716553,
      "epoch": 0.8472423802612482,
      "grad_norm": 12.0,
      "learning_rate": 0.000493559590061253,
      "loss": 5.6457,
      "mean_token_accuracy": 0.15094297528266906,
      "num_tokens": 19393692.0,
      "step": 9340
    },
    {
      "entropy": 5.887370777130127,
      "epoch": 0.8476959361393324,
      "grad_norm": 6.15625,
      "learning_rate": 0.0004935519016189559,
      "loss": 5.7067,
      "mean_token_accuracy": 0.15922312512993814,
      "num_tokens": 19404587.0,
      "step": 9345
    },
    {
      "entropy": 5.892665004730224,
      "epoch": 0.8481494920174165,
      "grad_norm": 3.375,
      "learning_rate": 0.0004935442086569498,
      "loss": 5.6613,
      "mean_token_accuracy": 0.15366343706846236,
      "num_tokens": 19415296.0,
      "step": 9350
    },
    {
      "entropy": 5.887033700942993,
      "epoch": 0.8486030478955007,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004935365111753939,
      "loss": 5.7199,
      "mean_token_accuracy": 0.15258205384016038,
      "num_tokens": 19426119.0,
      "step": 9355
    },
    {
      "entropy": 5.8707421779632565,
      "epoch": 0.8490566037735849,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004935288091744472,
      "loss": 5.6975,
      "mean_token_accuracy": 0.15224236696958543,
      "num_tokens": 19436828.0,
      "step": 9360
    },
    {
      "entropy": 5.894240617752075,
      "epoch": 0.8495101596516691,
      "grad_norm": 2.25,
      "learning_rate": 0.000493521102654269,
      "loss": 5.6845,
      "mean_token_accuracy": 0.14839867651462554,
      "num_tokens": 19446583.0,
      "step": 9365
    },
    {
      "entropy": 5.792936849594116,
      "epoch": 0.8499637155297532,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004935133916150187,
      "loss": 5.565,
      "mean_token_accuracy": 0.15941736847162247,
      "num_tokens": 19456757.0,
      "step": 9370
    },
    {
      "entropy": 5.754615211486817,
      "epoch": 0.8504172714078374,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004935056760568558,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1472718082368374,
      "num_tokens": 19467624.0,
      "step": 9375
    },
    {
      "entropy": 5.768962860107422,
      "epoch": 0.8508708272859217,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004934979559799399,
      "loss": 5.6101,
      "mean_token_accuracy": 0.15519556999206544,
      "num_tokens": 19477759.0,
      "step": 9380
    },
    {
      "entropy": 5.826801013946533,
      "epoch": 0.8513243831640058,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004934902313844306,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15790425688028337,
      "num_tokens": 19488877.0,
      "step": 9385
    },
    {
      "entropy": 5.80237717628479,
      "epoch": 0.85177793904209,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004934825022704875,
      "loss": 5.6248,
      "mean_token_accuracy": 0.15270906090736389,
      "num_tokens": 19499170.0,
      "step": 9390
    },
    {
      "entropy": 5.8249390602111815,
      "epoch": 0.8522314949201741,
      "grad_norm": 3.25,
      "learning_rate": 0.0004934747686382707,
      "loss": 5.6502,
      "mean_token_accuracy": 0.14975784569978715,
      "num_tokens": 19509944.0,
      "step": 9395
    },
    {
      "entropy": 5.788189125061035,
      "epoch": 0.8526850507982583,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004934670304879398,
      "loss": 5.5969,
      "mean_token_accuracy": 0.15632303059101105,
      "num_tokens": 19519014.0,
      "step": 9400
    },
    {
      "entropy": 5.77811369895935,
      "epoch": 0.8531386066763426,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004934592878196553,
      "loss": 5.6028,
      "mean_token_accuracy": 0.1545301377773285,
      "num_tokens": 19528586.0,
      "step": 9405
    },
    {
      "entropy": 5.871918487548828,
      "epoch": 0.8535921625544267,
      "grad_norm": 2.796875,
      "learning_rate": 0.000493451540633577,
      "loss": 5.7379,
      "mean_token_accuracy": 0.15096943005919455,
      "num_tokens": 19539912.0,
      "step": 9410
    },
    {
      "entropy": 5.886108541488648,
      "epoch": 0.8540457184325109,
      "grad_norm": 2.875,
      "learning_rate": 0.0004934437889298651,
      "loss": 5.6926,
      "mean_token_accuracy": 0.14735344499349595,
      "num_tokens": 19549860.0,
      "step": 9415
    },
    {
      "entropy": 5.843727397918701,
      "epoch": 0.854499274310595,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004934360327086799,
      "loss": 5.6329,
      "mean_token_accuracy": 0.1558349311351776,
      "num_tokens": 19559446.0,
      "step": 9420
    },
    {
      "entropy": 5.875371599197388,
      "epoch": 0.8549528301886793,
      "grad_norm": 3.578125,
      "learning_rate": 0.000493428271970182,
      "loss": 5.6847,
      "mean_token_accuracy": 0.15351752042770386,
      "num_tokens": 19570282.0,
      "step": 9425
    },
    {
      "entropy": 5.826085567474365,
      "epoch": 0.8554063860667634,
      "grad_norm": 3.0,
      "learning_rate": 0.0004934205067145318,
      "loss": 5.6183,
      "mean_token_accuracy": 0.14981367737054824,
      "num_tokens": 19581087.0,
      "step": 9430
    },
    {
      "entropy": 5.771932888031006,
      "epoch": 0.8558599419448476,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004934127369418898,
      "loss": 5.7064,
      "mean_token_accuracy": 0.15382656902074815,
      "num_tokens": 19592100.0,
      "step": 9435
    },
    {
      "entropy": 5.816186094284058,
      "epoch": 0.8563134978229318,
      "grad_norm": 4.0,
      "learning_rate": 0.0004934049626524168,
      "loss": 5.6612,
      "mean_token_accuracy": 0.15463659167289734,
      "num_tokens": 19602557.0,
      "step": 9440
    },
    {
      "entropy": 5.917910575866699,
      "epoch": 0.8567670537010159,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004933971838462734,
      "loss": 5.7029,
      "mean_token_accuracy": 0.14646178781986235,
      "num_tokens": 19613643.0,
      "step": 9445
    },
    {
      "entropy": 5.840442657470703,
      "epoch": 0.8572206095791002,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004933894005236207,
      "loss": 5.62,
      "mean_token_accuracy": 0.1540495663881302,
      "num_tokens": 19624180.0,
      "step": 9450
    },
    {
      "entropy": 5.812321996688842,
      "epoch": 0.8576741654571843,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004933816126846195,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1541966527700424,
      "num_tokens": 19634799.0,
      "step": 9455
    },
    {
      "entropy": 5.77674446105957,
      "epoch": 0.8581277213352685,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004933738203294309,
      "loss": 5.6135,
      "mean_token_accuracy": 0.160749851167202,
      "num_tokens": 19644618.0,
      "step": 9460
    },
    {
      "entropy": 5.760012865066528,
      "epoch": 0.8585812772133526,
      "grad_norm": 2.625,
      "learning_rate": 0.000493366023458216,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1502933233976364,
      "num_tokens": 19655550.0,
      "step": 9465
    },
    {
      "entropy": 5.8344627857208256,
      "epoch": 0.8590348330914369,
      "grad_norm": 3.25,
      "learning_rate": 0.0004933582220711362,
      "loss": 5.6693,
      "mean_token_accuracy": 0.15924992710351943,
      "num_tokens": 19665734.0,
      "step": 9470
    },
    {
      "entropy": 5.90078501701355,
      "epoch": 0.8594883889695211,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004933504161683526,
      "loss": 5.6556,
      "mean_token_accuracy": 0.14944824129343032,
      "num_tokens": 19677130.0,
      "step": 9475
    },
    {
      "entropy": 5.780390214920044,
      "epoch": 0.8599419448476052,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004933426057500269,
      "loss": 5.5754,
      "mean_token_accuracy": 0.15734326392412185,
      "num_tokens": 19687374.0,
      "step": 9480
    },
    {
      "entropy": 5.87744369506836,
      "epoch": 0.8603955007256894,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004933347908163203,
      "loss": 5.663,
      "mean_token_accuracy": 0.15029311627149583,
      "num_tokens": 19698182.0,
      "step": 9485
    },
    {
      "entropy": 5.847681093215942,
      "epoch": 0.8608490566037735,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004933269713673947,
      "loss": 5.6149,
      "mean_token_accuracy": 0.15695204436779023,
      "num_tokens": 19707664.0,
      "step": 9490
    },
    {
      "entropy": 5.727297925949097,
      "epoch": 0.8613026124818578,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004933191474034116,
      "loss": 5.6623,
      "mean_token_accuracy": 0.15112429186701776,
      "num_tokens": 19717647.0,
      "step": 9495
    },
    {
      "entropy": 5.788974571228027,
      "epoch": 0.861756168359942,
      "grad_norm": 2.484375,
      "learning_rate": 0.000493311318924533,
      "loss": 5.6965,
      "mean_token_accuracy": 0.15508863627910613,
      "num_tokens": 19727444.0,
      "step": 9500
    },
    {
      "entropy": 5.8454258918762205,
      "epoch": 0.8622097242380261,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004933034859309208,
      "loss": 5.7237,
      "mean_token_accuracy": 0.14808909147977828,
      "num_tokens": 19738719.0,
      "step": 9505
    },
    {
      "entropy": 5.907476949691772,
      "epoch": 0.8626632801161103,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004932956484227367,
      "loss": 5.6937,
      "mean_token_accuracy": 0.15322445929050446,
      "num_tokens": 19749786.0,
      "step": 9510
    },
    {
      "entropy": 5.898360776901245,
      "epoch": 0.8631168359941945,
      "grad_norm": 2.4375,
      "learning_rate": 0.000493287806400143,
      "loss": 5.677,
      "mean_token_accuracy": 0.16384984850883483,
      "num_tokens": 19759623.0,
      "step": 9515
    },
    {
      "entropy": 5.84280366897583,
      "epoch": 0.8635703918722787,
      "grad_norm": 2.75,
      "learning_rate": 0.0004932799598633018,
      "loss": 5.6466,
      "mean_token_accuracy": 0.1526094436645508,
      "num_tokens": 19770117.0,
      "step": 9520
    },
    {
      "entropy": 5.793286323547363,
      "epoch": 0.8640239477503628,
      "grad_norm": 3.125,
      "learning_rate": 0.0004932721088123753,
      "loss": 5.7419,
      "mean_token_accuracy": 0.15547346323728561,
      "num_tokens": 19780748.0,
      "step": 9525
    },
    {
      "entropy": 5.82646107673645,
      "epoch": 0.864477503628447,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004932642532475262,
      "loss": 5.712,
      "mean_token_accuracy": 0.1523199200630188,
      "num_tokens": 19791816.0,
      "step": 9530
    },
    {
      "entropy": 5.908600378036499,
      "epoch": 0.8649310595065312,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004932563931689168,
      "loss": 5.7278,
      "mean_token_accuracy": 0.15349765717983246,
      "num_tokens": 19801096.0,
      "step": 9535
    },
    {
      "entropy": 5.856309509277343,
      "epoch": 0.8653846153846154,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004932485285767093,
      "loss": 5.5903,
      "mean_token_accuracy": 0.1556379422545433,
      "num_tokens": 19811258.0,
      "step": 9540
    },
    {
      "entropy": 5.814109754562378,
      "epoch": 0.8658381712626996,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004932406594710667,
      "loss": 5.6841,
      "mean_token_accuracy": 0.15989995896816253,
      "num_tokens": 19821765.0,
      "step": 9545
    },
    {
      "entropy": 5.784682178497315,
      "epoch": 0.8662917271407837,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004932327858521517,
      "loss": 5.5602,
      "mean_token_accuracy": 0.16248162388801574,
      "num_tokens": 19831537.0,
      "step": 9550
    },
    {
      "entropy": 5.778852939605713,
      "epoch": 0.8667452830188679,
      "grad_norm": 2.71875,
      "learning_rate": 0.000493224907720127,
      "loss": 5.5746,
      "mean_token_accuracy": 0.15467148125171662,
      "num_tokens": 19841830.0,
      "step": 9555
    },
    {
      "entropy": 5.863460540771484,
      "epoch": 0.8671988388969522,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004932170250751556,
      "loss": 5.6807,
      "mean_token_accuracy": 0.1475910425186157,
      "num_tokens": 19851896.0,
      "step": 9560
    },
    {
      "entropy": 5.828677225112915,
      "epoch": 0.8676523947750363,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004932091379174006,
      "loss": 5.7099,
      "mean_token_accuracy": 0.14795950353145598,
      "num_tokens": 19861735.0,
      "step": 9565
    },
    {
      "entropy": 5.848706912994385,
      "epoch": 0.8681059506531205,
      "grad_norm": 2.71875,
      "learning_rate": 0.000493201246247025,
      "loss": 5.6144,
      "mean_token_accuracy": 0.1619333043694496,
      "num_tokens": 19872333.0,
      "step": 9570
    },
    {
      "entropy": 5.8216742992401125,
      "epoch": 0.8685595065312046,
      "grad_norm": 3.578125,
      "learning_rate": 0.000493193350064192,
      "loss": 5.6359,
      "mean_token_accuracy": 0.14962711334228515,
      "num_tokens": 19882896.0,
      "step": 9575
    },
    {
      "entropy": 5.810192537307739,
      "epoch": 0.8690130624092888,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004931854493690649,
      "loss": 5.5871,
      "mean_token_accuracy": 0.1537919670343399,
      "num_tokens": 19893767.0,
      "step": 9580
    },
    {
      "entropy": 5.7544848918914795,
      "epoch": 0.869466618287373,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004931775441618071,
      "loss": 5.5776,
      "mean_token_accuracy": 0.15696486085653305,
      "num_tokens": 19904242.0,
      "step": 9585
    },
    {
      "entropy": 5.792415761947632,
      "epoch": 0.8699201741654572,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004931696344425821,
      "loss": 5.6734,
      "mean_token_accuracy": 0.15541588217020036,
      "num_tokens": 19913898.0,
      "step": 9590
    },
    {
      "entropy": 5.925848960876465,
      "epoch": 0.8703737300435413,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004931617202115535,
      "loss": 5.6865,
      "mean_token_accuracy": 0.15690164417028427,
      "num_tokens": 19923756.0,
      "step": 9595
    },
    {
      "entropy": 5.770646429061889,
      "epoch": 0.8708272859216255,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004931538014688849,
      "loss": 5.6273,
      "mean_token_accuracy": 0.15638579279184342,
      "num_tokens": 19932720.0,
      "step": 9600
    },
    {
      "entropy": 5.847241497039795,
      "epoch": 0.8712808417997098,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004931458782147401,
      "loss": 5.8113,
      "mean_token_accuracy": 0.1438422456383705,
      "num_tokens": 19942962.0,
      "step": 9605
    },
    {
      "entropy": 5.976376676559449,
      "epoch": 0.8717343976777939,
      "grad_norm": 2.484375,
      "learning_rate": 0.000493137950449283,
      "loss": 5.6977,
      "mean_token_accuracy": 0.15650282204151153,
      "num_tokens": 19953735.0,
      "step": 9610
    },
    {
      "entropy": 5.831784582138061,
      "epoch": 0.8721879535558781,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004931300181726775,
      "loss": 5.6815,
      "mean_token_accuracy": 0.15415865927934647,
      "num_tokens": 19964416.0,
      "step": 9615
    },
    {
      "entropy": 5.827550315856934,
      "epoch": 0.8726415094339622,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004931220813850875,
      "loss": 5.6933,
      "mean_token_accuracy": 0.14974114447832107,
      "num_tokens": 19974531.0,
      "step": 9620
    },
    {
      "entropy": 5.856517553329468,
      "epoch": 0.8730950653120464,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004931141400866773,
      "loss": 5.6931,
      "mean_token_accuracy": 0.15448979288339615,
      "num_tokens": 19984964.0,
      "step": 9625
    },
    {
      "entropy": 5.804881381988525,
      "epoch": 0.8735486211901307,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004931061942776112,
      "loss": 5.6258,
      "mean_token_accuracy": 0.159735769033432,
      "num_tokens": 19995568.0,
      "step": 9630
    },
    {
      "entropy": 5.927798700332642,
      "epoch": 0.8740021770682148,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004930982439580534,
      "loss": 5.8032,
      "mean_token_accuracy": 0.14587000608444214,
      "num_tokens": 20005856.0,
      "step": 9635
    },
    {
      "entropy": 5.904760313034058,
      "epoch": 0.874455732946299,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004930902891281683,
      "loss": 5.6923,
      "mean_token_accuracy": 0.150040340423584,
      "num_tokens": 20015683.0,
      "step": 9640
    },
    {
      "entropy": 5.7723430633544925,
      "epoch": 0.8749092888243831,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004930823297881205,
      "loss": 5.534,
      "mean_token_accuracy": 0.15704142898321152,
      "num_tokens": 20025998.0,
      "step": 9645
    },
    {
      "entropy": 5.798841381072998,
      "epoch": 0.8753628447024674,
      "grad_norm": 9.5,
      "learning_rate": 0.0004930743659380745,
      "loss": 5.6927,
      "mean_token_accuracy": 0.15218583345413209,
      "num_tokens": 20036486.0,
      "step": 9650
    },
    {
      "entropy": 5.741021728515625,
      "epoch": 0.8758164005805515,
      "grad_norm": 2.859375,
      "learning_rate": 0.000493066397578195,
      "loss": 5.6093,
      "mean_token_accuracy": 0.15795097947120668,
      "num_tokens": 20046784.0,
      "step": 9655
    },
    {
      "entropy": 5.851765108108521,
      "epoch": 0.8762699564586357,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004930584247086468,
      "loss": 5.6552,
      "mean_token_accuracy": 0.15040991678833962,
      "num_tokens": 20056999.0,
      "step": 9660
    },
    {
      "entropy": 5.8147345066070555,
      "epoch": 0.8767235123367199,
      "grad_norm": 2.546875,
      "learning_rate": 0.000493050447329595,
      "loss": 5.5419,
      "mean_token_accuracy": 0.1588175803422928,
      "num_tokens": 20067612.0,
      "step": 9665
    },
    {
      "entropy": 5.856226444244385,
      "epoch": 0.877177068214804,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004930424654412043,
      "loss": 5.7058,
      "mean_token_accuracy": 0.1530055969953537,
      "num_tokens": 20077330.0,
      "step": 9670
    },
    {
      "entropy": 5.81877875328064,
      "epoch": 0.8776306240928883,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004930344790436398,
      "loss": 5.6697,
      "mean_token_accuracy": 0.1515201061964035,
      "num_tokens": 20086801.0,
      "step": 9675
    },
    {
      "entropy": 5.8394787311553955,
      "epoch": 0.8780841799709724,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004930264881370666,
      "loss": 5.6409,
      "mean_token_accuracy": 0.15190551653504372,
      "num_tokens": 20097047.0,
      "step": 9680
    },
    {
      "entropy": 5.822015190124512,
      "epoch": 0.8785377358490566,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004930184927216504,
      "loss": 5.6708,
      "mean_token_accuracy": 0.1601392611861229,
      "num_tokens": 20107185.0,
      "step": 9685
    },
    {
      "entropy": 5.829152631759643,
      "epoch": 0.8789912917271407,
      "grad_norm": 2.375,
      "learning_rate": 0.0004930104927975559,
      "loss": 5.6213,
      "mean_token_accuracy": 0.15499529242515564,
      "num_tokens": 20118493.0,
      "step": 9690
    },
    {
      "entropy": 5.790694379806519,
      "epoch": 0.879444847605225,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004930024883649489,
      "loss": 5.671,
      "mean_token_accuracy": 0.15504609644412995,
      "num_tokens": 20129294.0,
      "step": 9695
    },
    {
      "entropy": 5.7521278858184814,
      "epoch": 0.8798984034833092,
      "grad_norm": 2.421875,
      "learning_rate": 0.000492994479423995,
      "loss": 5.6159,
      "mean_token_accuracy": 0.1620168849825859,
      "num_tokens": 20139194.0,
      "step": 9700
    },
    {
      "entropy": 5.803879928588867,
      "epoch": 0.8803519593613933,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004929864659748597,
      "loss": 5.6527,
      "mean_token_accuracy": 0.1494958892464638,
      "num_tokens": 20149681.0,
      "step": 9705
    },
    {
      "entropy": 5.84139928817749,
      "epoch": 0.8808055152394775,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004929784480177087,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1462699994444847,
      "num_tokens": 20160557.0,
      "step": 9710
    },
    {
      "entropy": 5.8120523452758786,
      "epoch": 0.8812590711175616,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004929704255527078,
      "loss": 5.6344,
      "mean_token_accuracy": 0.15634542852640151,
      "num_tokens": 20171047.0,
      "step": 9715
    },
    {
      "entropy": 5.867328405380249,
      "epoch": 0.8817126269956459,
      "grad_norm": 3.28125,
      "learning_rate": 0.000492962398580023,
      "loss": 5.7423,
      "mean_token_accuracy": 0.15117280706763267,
      "num_tokens": 20182148.0,
      "step": 9720
    },
    {
      "entropy": 5.777922821044922,
      "epoch": 0.88216618287373,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004929543670998203,
      "loss": 5.5569,
      "mean_token_accuracy": 0.15868174731731416,
      "num_tokens": 20192131.0,
      "step": 9725
    },
    {
      "entropy": 5.855689859390258,
      "epoch": 0.8826197387518142,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004929463311122657,
      "loss": 5.6737,
      "mean_token_accuracy": 0.15011551976203918,
      "num_tokens": 20202805.0,
      "step": 9730
    },
    {
      "entropy": 5.817514896392822,
      "epoch": 0.8830732946298984,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004929382906175253,
      "loss": 5.6053,
      "mean_token_accuracy": 0.1565737932920456,
      "num_tokens": 20212702.0,
      "step": 9735
    },
    {
      "entropy": 5.780954742431641,
      "epoch": 0.8835268505079826,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004929302456157657,
      "loss": 5.6272,
      "mean_token_accuracy": 0.1581820070743561,
      "num_tokens": 20222661.0,
      "step": 9740
    },
    {
      "entropy": 5.840067434310913,
      "epoch": 0.8839804063860668,
      "grad_norm": 3.0,
      "learning_rate": 0.0004929221961071531,
      "loss": 5.6645,
      "mean_token_accuracy": 0.1551720142364502,
      "num_tokens": 20233012.0,
      "step": 9745
    },
    {
      "entropy": 5.858145475387573,
      "epoch": 0.8844339622641509,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004929141420918538,
      "loss": 5.6646,
      "mean_token_accuracy": 0.1597403571009636,
      "num_tokens": 20242918.0,
      "step": 9750
    },
    {
      "entropy": 5.802211093902588,
      "epoch": 0.8848875181422351,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004929060835700347,
      "loss": 5.6187,
      "mean_token_accuracy": 0.15960576385259628,
      "num_tokens": 20253254.0,
      "step": 9755
    },
    {
      "entropy": 5.8462591648101805,
      "epoch": 0.8853410740203193,
      "grad_norm": 2.546875,
      "learning_rate": 0.000492898020541862,
      "loss": 5.6733,
      "mean_token_accuracy": 0.14833195358514786,
      "num_tokens": 20263355.0,
      "step": 9760
    },
    {
      "entropy": 5.767037773132325,
      "epoch": 0.8857946298984035,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004928899530075029,
      "loss": 5.6089,
      "mean_token_accuracy": 0.1538158804178238,
      "num_tokens": 20274716.0,
      "step": 9765
    },
    {
      "entropy": 5.755797386169434,
      "epoch": 0.8862481857764877,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004928818809671241,
      "loss": 5.5942,
      "mean_token_accuracy": 0.15015506744384766,
      "num_tokens": 20285631.0,
      "step": 9770
    },
    {
      "entropy": 5.870469570159912,
      "epoch": 0.8867017416545718,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004928738044208923,
      "loss": 5.6521,
      "mean_token_accuracy": 0.15331775397062303,
      "num_tokens": 20295717.0,
      "step": 9775
    },
    {
      "entropy": 5.748063564300537,
      "epoch": 0.887155297532656,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004928657233689748,
      "loss": 5.6387,
      "mean_token_accuracy": 0.15630463659763336,
      "num_tokens": 20306637.0,
      "step": 9780
    },
    {
      "entropy": 5.818713998794555,
      "epoch": 0.8876088534107403,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004928576378115387,
      "loss": 5.5891,
      "mean_token_accuracy": 0.16262200921773912,
      "num_tokens": 20316629.0,
      "step": 9785
    },
    {
      "entropy": 5.896687126159668,
      "epoch": 0.8880624092888244,
      "grad_norm": 4.5,
      "learning_rate": 0.0004928495477487509,
      "loss": 5.788,
      "mean_token_accuracy": 0.14297403693199157,
      "num_tokens": 20327935.0,
      "step": 9790
    },
    {
      "entropy": 5.855598497390747,
      "epoch": 0.8885159651669086,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004928414531807792,
      "loss": 5.6003,
      "mean_token_accuracy": 0.1532953843474388,
      "num_tokens": 20339142.0,
      "step": 9795
    },
    {
      "entropy": 5.786568069458008,
      "epoch": 0.8889695210449927,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004928333541077906,
      "loss": 5.6035,
      "mean_token_accuracy": 0.16063882559537887,
      "num_tokens": 20348474.0,
      "step": 9800
    },
    {
      "entropy": 5.816909885406494,
      "epoch": 0.8894230769230769,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004928252505299527,
      "loss": 5.6577,
      "mean_token_accuracy": 0.1501513257622719,
      "num_tokens": 20359440.0,
      "step": 9805
    },
    {
      "entropy": 5.875770139694214,
      "epoch": 0.8898766328011611,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004928171424474332,
      "loss": 5.7049,
      "mean_token_accuracy": 0.15303952395915985,
      "num_tokens": 20370355.0,
      "step": 9810
    },
    {
      "entropy": 5.778997087478638,
      "epoch": 0.8903301886792453,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004928090298603996,
      "loss": 5.6012,
      "mean_token_accuracy": 0.15646144449710847,
      "num_tokens": 20380214.0,
      "step": 9815
    },
    {
      "entropy": 5.799949645996094,
      "epoch": 0.8907837445573294,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004928009127690199,
      "loss": 5.6068,
      "mean_token_accuracy": 0.15871121734380722,
      "num_tokens": 20389190.0,
      "step": 9820
    },
    {
      "entropy": 5.775676584243774,
      "epoch": 0.8912373004354136,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004927927911734617,
      "loss": 5.6645,
      "mean_token_accuracy": 0.15490778982639314,
      "num_tokens": 20399531.0,
      "step": 9825
    },
    {
      "entropy": 5.819215440750122,
      "epoch": 0.8916908563134979,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004927846650738931,
      "loss": 5.6684,
      "mean_token_accuracy": 0.15653486102819442,
      "num_tokens": 20410201.0,
      "step": 9830
    },
    {
      "entropy": 5.820386266708374,
      "epoch": 0.892144412191582,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004927765344704821,
      "loss": 5.6198,
      "mean_token_accuracy": 0.15861736387014388,
      "num_tokens": 20421579.0,
      "step": 9835
    },
    {
      "entropy": 5.789028739929199,
      "epoch": 0.8925979680696662,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004927683993633969,
      "loss": 5.6932,
      "mean_token_accuracy": 0.15348317474126816,
      "num_tokens": 20432048.0,
      "step": 9840
    },
    {
      "entropy": 5.8670917510986325,
      "epoch": 0.8930515239477503,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004927602597528057,
      "loss": 5.7223,
      "mean_token_accuracy": 0.15003873780369759,
      "num_tokens": 20441752.0,
      "step": 9845
    },
    {
      "entropy": 5.822627258300781,
      "epoch": 0.8935050798258345,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004927521156388769,
      "loss": 5.5655,
      "mean_token_accuracy": 0.15744317770004274,
      "num_tokens": 20451693.0,
      "step": 9850
    },
    {
      "entropy": 5.830567121505737,
      "epoch": 0.8939586357039188,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004927439670217788,
      "loss": 5.6907,
      "mean_token_accuracy": 0.1509220041334629,
      "num_tokens": 20462312.0,
      "step": 9855
    },
    {
      "entropy": 5.790587472915649,
      "epoch": 0.8944121915820029,
      "grad_norm": 2.625,
      "learning_rate": 0.00049273581390168,
      "loss": 5.6059,
      "mean_token_accuracy": 0.1467101141810417,
      "num_tokens": 20472546.0,
      "step": 9860
    },
    {
      "entropy": 5.781812763214111,
      "epoch": 0.8948657474600871,
      "grad_norm": 2.96875,
      "learning_rate": 0.000492727656278749,
      "loss": 5.6558,
      "mean_token_accuracy": 0.14958946257829667,
      "num_tokens": 20483266.0,
      "step": 9865
    },
    {
      "entropy": 5.750693082809448,
      "epoch": 0.8953193033381712,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004927194941531547,
      "loss": 5.5756,
      "mean_token_accuracy": 0.15675682723522186,
      "num_tokens": 20492490.0,
      "step": 9870
    },
    {
      "entropy": 5.833357667922973,
      "epoch": 0.8957728592162555,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004927113275250657,
      "loss": 5.6116,
      "mean_token_accuracy": 0.15225069373846054,
      "num_tokens": 20503077.0,
      "step": 9875
    },
    {
      "entropy": 5.757365465164185,
      "epoch": 0.8962264150943396,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004927031563946511,
      "loss": 5.6305,
      "mean_token_accuracy": 0.15087197571992875,
      "num_tokens": 20513771.0,
      "step": 9880
    },
    {
      "entropy": 5.80473780632019,
      "epoch": 0.8966799709724238,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004926949807620796,
      "loss": 5.7378,
      "mean_token_accuracy": 0.1536948338150978,
      "num_tokens": 20523616.0,
      "step": 9885
    },
    {
      "entropy": 5.843550491333008,
      "epoch": 0.897133526850508,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004926868006275204,
      "loss": 5.5529,
      "mean_token_accuracy": 0.15692562013864517,
      "num_tokens": 20533613.0,
      "step": 9890
    },
    {
      "entropy": 5.7657022953033445,
      "epoch": 0.8975870827285921,
      "grad_norm": 2.375,
      "learning_rate": 0.0004926786159911427,
      "loss": 5.6365,
      "mean_token_accuracy": 0.1459011420607567,
      "num_tokens": 20543812.0,
      "step": 9895
    },
    {
      "entropy": 5.727877378463745,
      "epoch": 0.8980406386066764,
      "grad_norm": 2.234375,
      "learning_rate": 0.000492670426853116,
      "loss": 5.5854,
      "mean_token_accuracy": 0.15662455111742019,
      "num_tokens": 20553785.0,
      "step": 9900
    },
    {
      "entropy": 5.887316608428955,
      "epoch": 0.8984941944847605,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004926622332136092,
      "loss": 5.6642,
      "mean_token_accuracy": 0.15611073970794678,
      "num_tokens": 20563326.0,
      "step": 9905
    },
    {
      "entropy": 5.789510202407837,
      "epoch": 0.8989477503628447,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004926540350727918,
      "loss": 5.5444,
      "mean_token_accuracy": 0.15752981156110762,
      "num_tokens": 20573643.0,
      "step": 9910
    },
    {
      "entropy": 5.686382246017456,
      "epoch": 0.8994013062409288,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004926458324308337,
      "loss": 5.5834,
      "mean_token_accuracy": 0.15729566514492035,
      "num_tokens": 20583916.0,
      "step": 9915
    },
    {
      "entropy": 5.836192798614502,
      "epoch": 0.8998548621190131,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004926376252879043,
      "loss": 5.7371,
      "mean_token_accuracy": 0.14930728375911712,
      "num_tokens": 20593984.0,
      "step": 9920
    },
    {
      "entropy": 5.861156177520752,
      "epoch": 0.9003084179970973,
      "grad_norm": 3.125,
      "learning_rate": 0.0004926294136441733,
      "loss": 5.5535,
      "mean_token_accuracy": 0.161505389213562,
      "num_tokens": 20603927.0,
      "step": 9925
    },
    {
      "entropy": 5.867801284790039,
      "epoch": 0.9007619738751814,
      "grad_norm": 6.1875,
      "learning_rate": 0.0004926211974998106,
      "loss": 5.7063,
      "mean_token_accuracy": 0.15205266922712327,
      "num_tokens": 20614147.0,
      "step": 9930
    },
    {
      "entropy": 5.894407272338867,
      "epoch": 0.9012155297532656,
      "grad_norm": 2.09375,
      "learning_rate": 0.000492612976854986,
      "loss": 5.6583,
      "mean_token_accuracy": 0.1546430841088295,
      "num_tokens": 20623631.0,
      "step": 9935
    },
    {
      "entropy": 5.720052576065063,
      "epoch": 0.9016690856313497,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004926047517098697,
      "loss": 5.5485,
      "mean_token_accuracy": 0.15941712707281114,
      "num_tokens": 20632566.0,
      "step": 9940
    },
    {
      "entropy": 5.809675502777099,
      "epoch": 0.902122641509434,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004925965220646315,
      "loss": 5.6289,
      "mean_token_accuracy": 0.15629922449588776,
      "num_tokens": 20643246.0,
      "step": 9945
    },
    {
      "entropy": 5.817538738250732,
      "epoch": 0.9025761973875182,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004925882879194419,
      "loss": 5.7432,
      "mean_token_accuracy": 0.1489126816391945,
      "num_tokens": 20653372.0,
      "step": 9950
    },
    {
      "entropy": 5.853244304656982,
      "epoch": 0.9030297532656023,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004925800492744712,
      "loss": 5.6704,
      "mean_token_accuracy": 0.15181304961442948,
      "num_tokens": 20662687.0,
      "step": 9955
    },
    {
      "entropy": 5.815691566467285,
      "epoch": 0.9034833091436865,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004925718061298895,
      "loss": 5.6074,
      "mean_token_accuracy": 0.16012194007635117,
      "num_tokens": 20673168.0,
      "step": 9960
    },
    {
      "entropy": 5.829687976837159,
      "epoch": 0.9039368650217707,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004925635584858675,
      "loss": 5.709,
      "mean_token_accuracy": 0.15368616580963135,
      "num_tokens": 20683100.0,
      "step": 9965
    },
    {
      "entropy": 5.898300218582153,
      "epoch": 0.9043904208998549,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004925553063425755,
      "loss": 5.7214,
      "mean_token_accuracy": 0.1502762645483017,
      "num_tokens": 20694097.0,
      "step": 9970
    },
    {
      "entropy": 5.8318281173706055,
      "epoch": 0.904843976777939,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004925470497001844,
      "loss": 5.6888,
      "mean_token_accuracy": 0.1505840465426445,
      "num_tokens": 20704640.0,
      "step": 9975
    },
    {
      "entropy": 5.79219069480896,
      "epoch": 0.9052975326560232,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004925387885588648,
      "loss": 5.6115,
      "mean_token_accuracy": 0.15379339158535005,
      "num_tokens": 20714419.0,
      "step": 9980
    },
    {
      "entropy": 5.930364370346069,
      "epoch": 0.9057510885341074,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004925305229187877,
      "loss": 5.7051,
      "mean_token_accuracy": 0.15184128955006598,
      "num_tokens": 20724705.0,
      "step": 9985
    },
    {
      "entropy": 5.943155097961426,
      "epoch": 0.9062046444121916,
      "grad_norm": 2.3125,
      "learning_rate": 0.000492522252780124,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1515795536339283,
      "num_tokens": 20734753.0,
      "step": 9990
    },
    {
      "entropy": 5.797655057907105,
      "epoch": 0.9066582002902758,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004925139781430446,
      "loss": 5.5928,
      "mean_token_accuracy": 0.156381855905056,
      "num_tokens": 20745389.0,
      "step": 9995
    },
    {
      "entropy": 5.760052299499511,
      "epoch": 0.9071117561683599,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004925056990077207,
      "loss": 5.6167,
      "mean_token_accuracy": 0.15757633447647096,
      "num_tokens": 20755702.0,
      "step": 10000
    },
    {
      "entropy": 5.840163516998291,
      "epoch": 0.9075653120464441,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004924974153743235,
      "loss": 5.5611,
      "mean_token_accuracy": 0.15907627418637277,
      "num_tokens": 20766046.0,
      "step": 10005
    },
    {
      "entropy": 5.841227865219116,
      "epoch": 0.9080188679245284,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004924891272430244,
      "loss": 5.6409,
      "mean_token_accuracy": 0.15797481536865235,
      "num_tokens": 20776654.0,
      "step": 10010
    },
    {
      "entropy": 5.900266551971436,
      "epoch": 0.9084724238026125,
      "grad_norm": 4.375,
      "learning_rate": 0.0004924808346139948,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1568849690258503,
      "num_tokens": 20787066.0,
      "step": 10015
    },
    {
      "entropy": 5.8268232345581055,
      "epoch": 0.9089259796806967,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004924725374874059,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15794997215270995,
      "num_tokens": 20796822.0,
      "step": 10020
    },
    {
      "entropy": 5.798414707183838,
      "epoch": 0.9093795355587808,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004924642358634297,
      "loss": 5.6485,
      "mean_token_accuracy": 0.15123971551656723,
      "num_tokens": 20807602.0,
      "step": 10025
    },
    {
      "entropy": 5.838248157501221,
      "epoch": 0.909833091436865,
      "grad_norm": 4.375,
      "learning_rate": 0.0004924559297422376,
      "loss": 5.6415,
      "mean_token_accuracy": 0.1583787500858307,
      "num_tokens": 20819046.0,
      "step": 10030
    },
    {
      "entropy": 5.831664276123047,
      "epoch": 0.9102866473149492,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004924476191240014,
      "loss": 5.6566,
      "mean_token_accuracy": 0.15371111333370208,
      "num_tokens": 20828989.0,
      "step": 10035
    },
    {
      "entropy": 5.811376571655273,
      "epoch": 0.9107402031930334,
      "grad_norm": 3.21875,
      "learning_rate": 0.000492439304008893,
      "loss": 5.6601,
      "mean_token_accuracy": 0.1526098906993866,
      "num_tokens": 20839366.0,
      "step": 10040
    },
    {
      "entropy": 5.830955839157104,
      "epoch": 0.9111937590711175,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004924309843970847,
      "loss": 5.6463,
      "mean_token_accuracy": 0.15849197879433632,
      "num_tokens": 20849513.0,
      "step": 10045
    },
    {
      "entropy": 5.910466909408569,
      "epoch": 0.9116473149492017,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004924226602887479,
      "loss": 5.7029,
      "mean_token_accuracy": 0.14674522280693053,
      "num_tokens": 20861406.0,
      "step": 10050
    },
    {
      "entropy": 5.799602031707764,
      "epoch": 0.912100870827286,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004924143316840552,
      "loss": 5.6015,
      "mean_token_accuracy": 0.16461957097053528,
      "num_tokens": 20871190.0,
      "step": 10055
    },
    {
      "entropy": 5.843008947372437,
      "epoch": 0.9125544267053701,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004924059985831787,
      "loss": 5.7387,
      "mean_token_accuracy": 0.15071364492177963,
      "num_tokens": 20881589.0,
      "step": 10060
    },
    {
      "entropy": 5.8244470119476315,
      "epoch": 0.9130079825834543,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004923976609862908,
      "loss": 5.6717,
      "mean_token_accuracy": 0.14722735732793807,
      "num_tokens": 20892072.0,
      "step": 10065
    },
    {
      "entropy": 5.886385250091553,
      "epoch": 0.9134615384615384,
      "grad_norm": 2.125,
      "learning_rate": 0.0004923893188935638,
      "loss": 5.6589,
      "mean_token_accuracy": 0.15132659822702407,
      "num_tokens": 20902616.0,
      "step": 10070
    },
    {
      "entropy": 5.869722700119018,
      "epoch": 0.9139150943396226,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004923809723051702,
      "loss": 5.6827,
      "mean_token_accuracy": 0.1534601032733917,
      "num_tokens": 20913832.0,
      "step": 10075
    },
    {
      "entropy": 5.739912796020508,
      "epoch": 0.9143686502177069,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004923726212212829,
      "loss": 5.5464,
      "mean_token_accuracy": 0.16056860238313675,
      "num_tokens": 20923322.0,
      "step": 10080
    },
    {
      "entropy": 5.758353137969971,
      "epoch": 0.914822206095791,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004923642656420742,
      "loss": 5.6584,
      "mean_token_accuracy": 0.15374343991279601,
      "num_tokens": 20933334.0,
      "step": 10085
    },
    {
      "entropy": 5.862091779708862,
      "epoch": 0.9152757619738752,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004923559055677173,
      "loss": 5.6023,
      "mean_token_accuracy": 0.1538617268204689,
      "num_tokens": 20944171.0,
      "step": 10090
    },
    {
      "entropy": 5.853946924209595,
      "epoch": 0.9157293178519593,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004923475409983848,
      "loss": 5.6513,
      "mean_token_accuracy": 0.14951413869857788,
      "num_tokens": 20954200.0,
      "step": 10095
    },
    {
      "entropy": 5.786560821533203,
      "epoch": 0.9161828737300436,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004923391719342497,
      "loss": 5.6525,
      "mean_token_accuracy": 0.1497160390019417,
      "num_tokens": 20964935.0,
      "step": 10100
    },
    {
      "entropy": 5.829623651504517,
      "epoch": 0.9166364296081277,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004923307983754852,
      "loss": 5.651,
      "mean_token_accuracy": 0.15524252504110336,
      "num_tokens": 20975499.0,
      "step": 10105
    },
    {
      "entropy": 5.718609046936035,
      "epoch": 0.9170899854862119,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004923224203222645,
      "loss": 5.5592,
      "mean_token_accuracy": 0.1539004072546959,
      "num_tokens": 20986199.0,
      "step": 10110
    },
    {
      "entropy": 5.755342245101929,
      "epoch": 0.9175435413642961,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004923140377747606,
      "loss": 5.6396,
      "mean_token_accuracy": 0.15343184471130372,
      "num_tokens": 20996313.0,
      "step": 10115
    },
    {
      "entropy": 5.819121789932251,
      "epoch": 0.9179970972423802,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004923056507331471,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1543293446302414,
      "num_tokens": 21006133.0,
      "step": 10120
    },
    {
      "entropy": 5.8339330673217775,
      "epoch": 0.9184506531204645,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004922972591975974,
      "loss": 5.5978,
      "mean_token_accuracy": 0.15862650871276857,
      "num_tokens": 21016448.0,
      "step": 10125
    },
    {
      "entropy": 5.770276260375977,
      "epoch": 0.9189042089985486,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004922888631682849,
      "loss": 5.5403,
      "mean_token_accuracy": 0.15439814180135727,
      "num_tokens": 21026458.0,
      "step": 10130
    },
    {
      "entropy": 5.825389194488525,
      "epoch": 0.9193577648766328,
      "grad_norm": 2.375,
      "learning_rate": 0.0004922804626453834,
      "loss": 5.5758,
      "mean_token_accuracy": 0.15657375603914261,
      "num_tokens": 21037111.0,
      "step": 10135
    },
    {
      "entropy": 5.813408279418946,
      "epoch": 0.9198113207547169,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004922720576290665,
      "loss": 5.6685,
      "mean_token_accuracy": 0.15609802007675172,
      "num_tokens": 21048041.0,
      "step": 10140
    },
    {
      "entropy": 5.73240475654602,
      "epoch": 0.9202648766328012,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004922636481195082,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16573669612407685,
      "num_tokens": 21058099.0,
      "step": 10145
    },
    {
      "entropy": 5.816942262649536,
      "epoch": 0.9207184325108854,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004922552341168823,
      "loss": 5.6739,
      "mean_token_accuracy": 0.14541862457990645,
      "num_tokens": 21068137.0,
      "step": 10150
    },
    {
      "entropy": 5.747505617141724,
      "epoch": 0.9211719883889695,
      "grad_norm": 2.625,
      "learning_rate": 0.0004922468156213627,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1605380281805992,
      "num_tokens": 21077556.0,
      "step": 10155
    },
    {
      "entropy": 5.804719352722168,
      "epoch": 0.9216255442670537,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004922383926331236,
      "loss": 5.5838,
      "mean_token_accuracy": 0.15588331669569017,
      "num_tokens": 21087524.0,
      "step": 10160
    },
    {
      "entropy": 5.7460955619812015,
      "epoch": 0.9220791001451378,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004922299651523392,
      "loss": 5.6607,
      "mean_token_accuracy": 0.15176035612821578,
      "num_tokens": 21098446.0,
      "step": 10165
    },
    {
      "entropy": 5.776041746139526,
      "epoch": 0.9225326560232221,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004922215331791837,
      "loss": 5.6514,
      "mean_token_accuracy": 0.1542374908924103,
      "num_tokens": 21109291.0,
      "step": 10170
    },
    {
      "entropy": 5.798018217086792,
      "epoch": 0.9229862119013063,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004922130967138316,
      "loss": 5.5424,
      "mean_token_accuracy": 0.1597289949655533,
      "num_tokens": 21120043.0,
      "step": 10175
    },
    {
      "entropy": 5.842937231063843,
      "epoch": 0.9234397677793904,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004922046557564574,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1505768433213234,
      "num_tokens": 21130339.0,
      "step": 10180
    },
    {
      "entropy": 5.839677381515503,
      "epoch": 0.9238933236574746,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004921962103072355,
      "loss": 5.7313,
      "mean_token_accuracy": 0.15170611292123795,
      "num_tokens": 21142307.0,
      "step": 10185
    },
    {
      "entropy": 5.841834831237793,
      "epoch": 0.9243468795355588,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004921877603663406,
      "loss": 5.6285,
      "mean_token_accuracy": 0.15684072822332382,
      "num_tokens": 21151843.0,
      "step": 10190
    },
    {
      "entropy": 5.84149398803711,
      "epoch": 0.924800435413643,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004921793059339474,
      "loss": 5.6204,
      "mean_token_accuracy": 0.1543716825544834,
      "num_tokens": 21162574.0,
      "step": 10195
    },
    {
      "entropy": 5.825618267059326,
      "epoch": 0.9252539912917271,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004921708470102309,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15569959282875062,
      "num_tokens": 21172637.0,
      "step": 10200
    },
    {
      "entropy": 5.734502506256104,
      "epoch": 0.9257075471698113,
      "grad_norm": 1.734375,
      "learning_rate": 0.000492162383595366,
      "loss": 5.5819,
      "mean_token_accuracy": 0.15843692421913147,
      "num_tokens": 21182272.0,
      "step": 10205
    },
    {
      "entropy": 5.8643608570098875,
      "epoch": 0.9261611030478955,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004921539156895275,
      "loss": 5.7286,
      "mean_token_accuracy": 0.15053602457046508,
      "num_tokens": 21193712.0,
      "step": 10210
    },
    {
      "entropy": 5.8853858470916744,
      "epoch": 0.9266146589259797,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004921454432928908,
      "loss": 5.718,
      "mean_token_accuracy": 0.14995245784521102,
      "num_tokens": 21205104.0,
      "step": 10215
    },
    {
      "entropy": 5.846768569946289,
      "epoch": 0.9270682148040639,
      "grad_norm": 2.578125,
      "learning_rate": 0.000492136966405631,
      "loss": 5.6111,
      "mean_token_accuracy": 0.15548471063375474,
      "num_tokens": 21215495.0,
      "step": 10220
    },
    {
      "entropy": 5.789591693878174,
      "epoch": 0.927521770682148,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004921284850279233,
      "loss": 5.6316,
      "mean_token_accuracy": 0.15263270437717438,
      "num_tokens": 21226047.0,
      "step": 10225
    },
    {
      "entropy": 5.819576454162598,
      "epoch": 0.9279753265602322,
      "grad_norm": 3.125,
      "learning_rate": 0.0004921199991599433,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1504075489938259,
      "num_tokens": 21235784.0,
      "step": 10230
    },
    {
      "entropy": 5.873410987854004,
      "epoch": 0.9284288824383164,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004921115088018665,
      "loss": 5.6496,
      "mean_token_accuracy": 0.15487141907215118,
      "num_tokens": 21245493.0,
      "step": 10235
    },
    {
      "entropy": 5.891049909591675,
      "epoch": 0.9288824383164006,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000492103013953868,
      "loss": 5.7419,
      "mean_token_accuracy": 0.14885285943746568,
      "num_tokens": 21255541.0,
      "step": 10240
    },
    {
      "entropy": 5.756620264053344,
      "epoch": 0.9293359941944848,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004920945146161242,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1582723706960678,
      "num_tokens": 21265427.0,
      "step": 10245
    },
    {
      "entropy": 5.780202388763428,
      "epoch": 0.9297895500725689,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004920860107888105,
      "loss": 5.605,
      "mean_token_accuracy": 0.1546992763876915,
      "num_tokens": 21276597.0,
      "step": 10250
    },
    {
      "entropy": 5.783038330078125,
      "epoch": 0.9302431059506531,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004920775024721026,
      "loss": 5.604,
      "mean_token_accuracy": 0.1555513247847557,
      "num_tokens": 21286416.0,
      "step": 10255
    },
    {
      "entropy": 5.770952939987183,
      "epoch": 0.9306966618287373,
      "grad_norm": 2.25,
      "learning_rate": 0.0004920689896661767,
      "loss": 5.6224,
      "mean_token_accuracy": 0.16034480333328247,
      "num_tokens": 21296064.0,
      "step": 10260
    },
    {
      "entropy": 5.796384191513061,
      "epoch": 0.9311502177068215,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004920604723712088,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1589336574077606,
      "num_tokens": 21306836.0,
      "step": 10265
    },
    {
      "entropy": 5.765583276748657,
      "epoch": 0.9316037735849056,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004920519505873751,
      "loss": 5.5942,
      "mean_token_accuracy": 0.1574314147233963,
      "num_tokens": 21316936.0,
      "step": 10270
    },
    {
      "entropy": 5.7488853454589846,
      "epoch": 0.9320573294629898,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004920434243148515,
      "loss": 5.5772,
      "mean_token_accuracy": 0.16025102734565735,
      "num_tokens": 21326892.0,
      "step": 10275
    },
    {
      "entropy": 5.79693112373352,
      "epoch": 0.9325108853410741,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004920348935538147,
      "loss": 5.5968,
      "mean_token_accuracy": 0.15238452702760696,
      "num_tokens": 21337270.0,
      "step": 10280
    },
    {
      "entropy": 5.816615438461303,
      "epoch": 0.9329644412191582,
      "grad_norm": 2.21875,
      "learning_rate": 0.000492026358304441,
      "loss": 5.664,
      "mean_token_accuracy": 0.15377061069011688,
      "num_tokens": 21347968.0,
      "step": 10285
    },
    {
      "entropy": 5.80585355758667,
      "epoch": 0.9334179970972424,
      "grad_norm": 2.578125,
      "learning_rate": 0.000492017818566907,
      "loss": 5.6084,
      "mean_token_accuracy": 0.1599868893623352,
      "num_tokens": 21358312.0,
      "step": 10290
    },
    {
      "entropy": 5.83113660812378,
      "epoch": 0.9338715529753265,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004920092743413892,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15949864387512208,
      "num_tokens": 21368794.0,
      "step": 10295
    },
    {
      "entropy": 5.85403733253479,
      "epoch": 0.9343251088534107,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004920007256280642,
      "loss": 5.6545,
      "mean_token_accuracy": 0.1595981612801552,
      "num_tokens": 21380080.0,
      "step": 10300
    },
    {
      "entropy": 5.783421802520752,
      "epoch": 0.934778664731495,
      "grad_norm": 2.296875,
      "learning_rate": 0.000491992172427109,
      "loss": 5.6109,
      "mean_token_accuracy": 0.15633728057146073,
      "num_tokens": 21389961.0,
      "step": 10305
    },
    {
      "entropy": 5.724167251586914,
      "epoch": 0.9352322206095791,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004919836147387002,
      "loss": 5.6492,
      "mean_token_accuracy": 0.15565934926271438,
      "num_tokens": 21399869.0,
      "step": 10310
    },
    {
      "entropy": 5.795440196990967,
      "epoch": 0.9356857764876633,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004919750525630151,
      "loss": 5.6292,
      "mean_token_accuracy": 0.15735144168138504,
      "num_tokens": 21410343.0,
      "step": 10315
    },
    {
      "entropy": 5.877757787704468,
      "epoch": 0.9361393323657474,
      "grad_norm": 2.75,
      "learning_rate": 0.0004919664859002306,
      "loss": 5.6479,
      "mean_token_accuracy": 0.1598845586180687,
      "num_tokens": 21420915.0,
      "step": 10320
    },
    {
      "entropy": 5.834699201583862,
      "epoch": 0.9365928882438317,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004919579147505239,
      "loss": 5.6066,
      "mean_token_accuracy": 0.1559985712170601,
      "num_tokens": 21430879.0,
      "step": 10325
    },
    {
      "entropy": 5.776730012893677,
      "epoch": 0.9370464441219158,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004919493391140721,
      "loss": 5.6739,
      "mean_token_accuracy": 0.15353767275810243,
      "num_tokens": 21440843.0,
      "step": 10330
    },
    {
      "entropy": 5.837963724136353,
      "epoch": 0.9375,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004919407589910529,
      "loss": 5.8305,
      "mean_token_accuracy": 0.14749686270952225,
      "num_tokens": 21452132.0,
      "step": 10335
    },
    {
      "entropy": 5.822698879241943,
      "epoch": 0.9379535558780842,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004919321743816434,
      "loss": 5.6264,
      "mean_token_accuracy": 0.15394114553928376,
      "num_tokens": 21461861.0,
      "step": 10340
    },
    {
      "entropy": 5.831479740142822,
      "epoch": 0.9384071117561683,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004919235852860213,
      "loss": 5.626,
      "mean_token_accuracy": 0.15050584822893143,
      "num_tokens": 21472244.0,
      "step": 10345
    },
    {
      "entropy": 5.858907270431518,
      "epoch": 0.9388606676342526,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004919149917043641,
      "loss": 5.6769,
      "mean_token_accuracy": 0.15351335778832437,
      "num_tokens": 21482394.0,
      "step": 10350
    },
    {
      "entropy": 5.798162841796875,
      "epoch": 0.9393142235123367,
      "grad_norm": 3.5,
      "learning_rate": 0.0004919063936368496,
      "loss": 5.724,
      "mean_token_accuracy": 0.1505119249224663,
      "num_tokens": 21493334.0,
      "step": 10355
    },
    {
      "entropy": 5.903670835494995,
      "epoch": 0.9397677793904209,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004918977910836557,
      "loss": 5.6716,
      "mean_token_accuracy": 0.15008047968149185,
      "num_tokens": 21504443.0,
      "step": 10360
    },
    {
      "entropy": 5.919463634490967,
      "epoch": 0.940221335268505,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004918891840449601,
      "loss": 5.7001,
      "mean_token_accuracy": 0.1535200819373131,
      "num_tokens": 21513908.0,
      "step": 10365
    },
    {
      "entropy": 5.819669914245606,
      "epoch": 0.9406748911465893,
      "grad_norm": 2.515625,
      "learning_rate": 0.000491880572520941,
      "loss": 5.7129,
      "mean_token_accuracy": 0.15204098969697952,
      "num_tokens": 21524349.0,
      "step": 10370
    },
    {
      "entropy": 5.970065927505493,
      "epoch": 0.9411284470246735,
      "grad_norm": 1.75,
      "learning_rate": 0.0004918719565117763,
      "loss": 5.7562,
      "mean_token_accuracy": 0.1551732286810875,
      "num_tokens": 21535003.0,
      "step": 10375
    },
    {
      "entropy": 5.940258646011353,
      "epoch": 0.9415820029027576,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004918633360176444,
      "loss": 5.8002,
      "mean_token_accuracy": 0.1469562128186226,
      "num_tokens": 21545045.0,
      "step": 10380
    },
    {
      "entropy": 5.784483289718628,
      "epoch": 0.9420355587808418,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004918547110387234,
      "loss": 5.6039,
      "mean_token_accuracy": 0.15595303624868392,
      "num_tokens": 21554830.0,
      "step": 10385
    },
    {
      "entropy": 5.826089000701904,
      "epoch": 0.9424891146589259,
      "grad_norm": 3.625,
      "learning_rate": 0.0004918460815751916,
      "loss": 5.5774,
      "mean_token_accuracy": 0.15867999494075774,
      "num_tokens": 21564693.0,
      "step": 10390
    },
    {
      "entropy": 5.778178453445435,
      "epoch": 0.9429426705370102,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004918374476272277,
      "loss": 5.5948,
      "mean_token_accuracy": 0.15591406226158142,
      "num_tokens": 21575475.0,
      "step": 10395
    },
    {
      "entropy": 5.782553958892822,
      "epoch": 0.9433962264150944,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004918288091950101,
      "loss": 5.6214,
      "mean_token_accuracy": 0.15539320111274718,
      "num_tokens": 21586944.0,
      "step": 10400
    },
    {
      "entropy": 5.872455310821533,
      "epoch": 0.9438497822931785,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004918201662787174,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1531512513756752,
      "num_tokens": 21598170.0,
      "step": 10405
    },
    {
      "entropy": 5.838468408584594,
      "epoch": 0.9443033381712627,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004918115188785285,
      "loss": 5.6587,
      "mean_token_accuracy": 0.15409923940896988,
      "num_tokens": 21609170.0,
      "step": 10410
    },
    {
      "entropy": 5.841028451919556,
      "epoch": 0.9447568940493469,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004918028669946221,
      "loss": 5.7217,
      "mean_token_accuracy": 0.15151064842939377,
      "num_tokens": 21620171.0,
      "step": 10415
    },
    {
      "entropy": 5.8394111633300785,
      "epoch": 0.9452104499274311,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004917942106271771,
      "loss": 5.6671,
      "mean_token_accuracy": 0.16137124970555305,
      "num_tokens": 21631072.0,
      "step": 10420
    },
    {
      "entropy": 5.75335693359375,
      "epoch": 0.9456640058055152,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004917855497763727,
      "loss": 5.5325,
      "mean_token_accuracy": 0.16007351726293564,
      "num_tokens": 21641543.0,
      "step": 10425
    },
    {
      "entropy": 5.783782625198365,
      "epoch": 0.9461175616835994,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004917768844423878,
      "loss": 5.6278,
      "mean_token_accuracy": 0.15471049398183823,
      "num_tokens": 21651342.0,
      "step": 10430
    },
    {
      "entropy": 5.888284301757812,
      "epoch": 0.9465711175616836,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004917682146254017,
      "loss": 5.6243,
      "mean_token_accuracy": 0.16156017035245895,
      "num_tokens": 21661043.0,
      "step": 10435
    },
    {
      "entropy": 5.774037456512451,
      "epoch": 0.9470246734397678,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004917595403255938,
      "loss": 5.4839,
      "mean_token_accuracy": 0.1669143870472908,
      "num_tokens": 21671197.0,
      "step": 10440
    },
    {
      "entropy": 5.731167411804199,
      "epoch": 0.947478229317852,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004917508615431431,
      "loss": 5.5763,
      "mean_token_accuracy": 0.15985903590917588,
      "num_tokens": 21682066.0,
      "step": 10445
    },
    {
      "entropy": 5.862526941299438,
      "epoch": 0.9479317851959361,
      "grad_norm": 2.25,
      "learning_rate": 0.0004917421782782296,
      "loss": 5.6135,
      "mean_token_accuracy": 0.15685460567474366,
      "num_tokens": 21692117.0,
      "step": 10450
    },
    {
      "entropy": 5.820763969421387,
      "epoch": 0.9483853410740203,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004917334905310325,
      "loss": 5.7187,
      "mean_token_accuracy": 0.1483694888651371,
      "num_tokens": 21703544.0,
      "step": 10455
    },
    {
      "entropy": 5.784500217437744,
      "epoch": 0.9488388969521045,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004917247983017317,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1575774610042572,
      "num_tokens": 21712850.0,
      "step": 10460
    },
    {
      "entropy": 5.857076168060303,
      "epoch": 0.9492924528301887,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004917161015905067,
      "loss": 5.7064,
      "mean_token_accuracy": 0.1546853482723236,
      "num_tokens": 21723015.0,
      "step": 10465
    },
    {
      "entropy": 5.780153369903564,
      "epoch": 0.9497460087082729,
      "grad_norm": 1.875,
      "learning_rate": 0.0004917074003975376,
      "loss": 5.6598,
      "mean_token_accuracy": 0.15189022123813628,
      "num_tokens": 21732453.0,
      "step": 10470
    },
    {
      "entropy": 5.802216529846191,
      "epoch": 0.950199564586357,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004916986947230042,
      "loss": 5.7187,
      "mean_token_accuracy": 0.15118529945611953,
      "num_tokens": 21743251.0,
      "step": 10475
    },
    {
      "entropy": 5.887509822845459,
      "epoch": 0.9506531204644412,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004916899845670866,
      "loss": 5.6239,
      "mean_token_accuracy": 0.15586574971675873,
      "num_tokens": 21752891.0,
      "step": 10480
    },
    {
      "entropy": 5.838762617111206,
      "epoch": 0.9511066763425254,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004916812699299649,
      "loss": 5.6682,
      "mean_token_accuracy": 0.15595989674329758,
      "num_tokens": 21763230.0,
      "step": 10485
    },
    {
      "entropy": 5.8394464492797855,
      "epoch": 0.9515602322206096,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004916725508118193,
      "loss": 5.5997,
      "mean_token_accuracy": 0.15375525653362274,
      "num_tokens": 21773663.0,
      "step": 10490
    },
    {
      "entropy": 5.81770887374878,
      "epoch": 0.9520137880986937,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004916638272128301,
      "loss": 5.6637,
      "mean_token_accuracy": 0.15654143393039704,
      "num_tokens": 21784265.0,
      "step": 10495
    },
    {
      "entropy": 5.745655345916748,
      "epoch": 0.9524673439767779,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004916550991331777,
      "loss": 5.6368,
      "mean_token_accuracy": 0.15793596804142,
      "num_tokens": 21795056.0,
      "step": 10500
    },
    {
      "entropy": 5.8132641315460205,
      "epoch": 0.9529208998548622,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004916463665730427,
      "loss": 5.6342,
      "mean_token_accuracy": 0.15889952480793,
      "num_tokens": 21806204.0,
      "step": 10505
    },
    {
      "entropy": 5.8509033679962155,
      "epoch": 0.9533744557329463,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004916376295326057,
      "loss": 5.6527,
      "mean_token_accuracy": 0.15502676367759705,
      "num_tokens": 21816993.0,
      "step": 10510
    },
    {
      "entropy": 5.824394702911377,
      "epoch": 0.9538280116110305,
      "grad_norm": 2.75,
      "learning_rate": 0.0004916288880120472,
      "loss": 5.5126,
      "mean_token_accuracy": 0.16500263810157775,
      "num_tokens": 21827457.0,
      "step": 10515
    },
    {
      "entropy": 5.768416929244995,
      "epoch": 0.9542815674891146,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004916201420115482,
      "loss": 5.608,
      "mean_token_accuracy": 0.15698783695697785,
      "num_tokens": 21837555.0,
      "step": 10520
    },
    {
      "entropy": 5.842128801345825,
      "epoch": 0.9547351233671988,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004916113915312895,
      "loss": 5.6454,
      "mean_token_accuracy": 0.16147412061691285,
      "num_tokens": 21847074.0,
      "step": 10525
    },
    {
      "entropy": 5.8470460891723635,
      "epoch": 0.9551886792452831,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004916026365714519,
      "loss": 5.642,
      "mean_token_accuracy": 0.1498659148812294,
      "num_tokens": 21858333.0,
      "step": 10530
    },
    {
      "entropy": 5.770013284683228,
      "epoch": 0.9556422351233672,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004915938771322165,
      "loss": 5.653,
      "mean_token_accuracy": 0.15716380327939988,
      "num_tokens": 21869132.0,
      "step": 10535
    },
    {
      "entropy": 5.758356142044067,
      "epoch": 0.9560957910014514,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004915851132137646,
      "loss": 5.6858,
      "mean_token_accuracy": 0.15203768908977508,
      "num_tokens": 21879153.0,
      "step": 10540
    },
    {
      "entropy": 5.845894527435303,
      "epoch": 0.9565493468795355,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004915763448162774,
      "loss": 5.5592,
      "mean_token_accuracy": 0.15751828402280807,
      "num_tokens": 21889063.0,
      "step": 10545
    },
    {
      "entropy": 5.8422064781188965,
      "epoch": 0.9570029027576198,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004915675719399362,
      "loss": 5.6833,
      "mean_token_accuracy": 0.15525020360946656,
      "num_tokens": 21899005.0,
      "step": 10550
    },
    {
      "entropy": 5.801091575622559,
      "epoch": 0.9574564586357039,
      "grad_norm": 3.0,
      "learning_rate": 0.0004915587945849225,
      "loss": 5.5429,
      "mean_token_accuracy": 0.15986981689929963,
      "num_tokens": 21908391.0,
      "step": 10555
    },
    {
      "entropy": 5.783149099349975,
      "epoch": 0.9579100145137881,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004915500127514176,
      "loss": 5.626,
      "mean_token_accuracy": 0.1556545153260231,
      "num_tokens": 21917655.0,
      "step": 10560
    },
    {
      "entropy": 5.848690319061279,
      "epoch": 0.9583635703918723,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004915412264396034,
      "loss": 5.6832,
      "mean_token_accuracy": 0.1582992359995842,
      "num_tokens": 21928038.0,
      "step": 10565
    },
    {
      "entropy": 5.786609077453614,
      "epoch": 0.9588171262699564,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004915324356496613,
      "loss": 5.5746,
      "mean_token_accuracy": 0.16353203654289244,
      "num_tokens": 21938106.0,
      "step": 10570
    },
    {
      "entropy": 5.773234176635742,
      "epoch": 0.9592706821480407,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004915236403817733,
      "loss": 5.5702,
      "mean_token_accuracy": 0.16138877272605895,
      "num_tokens": 21947484.0,
      "step": 10575
    },
    {
      "entropy": 5.814287567138672,
      "epoch": 0.9597242380261248,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004915148406361213,
      "loss": 5.6499,
      "mean_token_accuracy": 0.15784090012311935,
      "num_tokens": 21957139.0,
      "step": 10580
    },
    {
      "entropy": 5.769003677368164,
      "epoch": 0.960177793904209,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004915060364128872,
      "loss": 5.6391,
      "mean_token_accuracy": 0.14913761019706726,
      "num_tokens": 21967030.0,
      "step": 10585
    },
    {
      "entropy": 5.797343826293945,
      "epoch": 0.9606313497822931,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004914972277122532,
      "loss": 5.614,
      "mean_token_accuracy": 0.15193152502179147,
      "num_tokens": 21976418.0,
      "step": 10590
    },
    {
      "entropy": 5.77819242477417,
      "epoch": 0.9610849056603774,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004914884145344012,
      "loss": 5.6184,
      "mean_token_accuracy": 0.16035711616277695,
      "num_tokens": 21987284.0,
      "step": 10595
    },
    {
      "entropy": 5.716995763778686,
      "epoch": 0.9615384615384616,
      "grad_norm": 2.25,
      "learning_rate": 0.0004914795968795137,
      "loss": 5.5251,
      "mean_token_accuracy": 0.16215120255947113,
      "num_tokens": 21997106.0,
      "step": 10600
    },
    {
      "entropy": 5.741790771484375,
      "epoch": 0.9619920174165457,
      "grad_norm": 2.625,
      "learning_rate": 0.0004914707747477728,
      "loss": 5.539,
      "mean_token_accuracy": 0.1570558652281761,
      "num_tokens": 22007589.0,
      "step": 10605
    },
    {
      "entropy": 5.734328365325927,
      "epoch": 0.9624455732946299,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004914619481393613,
      "loss": 5.6412,
      "mean_token_accuracy": 0.15544965565204621,
      "num_tokens": 22017669.0,
      "step": 10610
    },
    {
      "entropy": 5.949207544326782,
      "epoch": 0.962899129172714,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004914531170544615,
      "loss": 5.8122,
      "mean_token_accuracy": 0.14744702577590943,
      "num_tokens": 22029330.0,
      "step": 10615
    },
    {
      "entropy": 5.87768325805664,
      "epoch": 0.9633526850507983,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004914442814932561,
      "loss": 5.5873,
      "mean_token_accuracy": 0.15414217859506607,
      "num_tokens": 22039877.0,
      "step": 10620
    },
    {
      "entropy": 5.795701313018799,
      "epoch": 0.9638062409288825,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004914354414559278,
      "loss": 5.6064,
      "mean_token_accuracy": 0.15927164107561112,
      "num_tokens": 22049579.0,
      "step": 10625
    },
    {
      "entropy": 5.655880784988403,
      "epoch": 0.9642597968069666,
      "grad_norm": 2.125,
      "learning_rate": 0.0004914265969426594,
      "loss": 5.5232,
      "mean_token_accuracy": 0.1695343554019928,
      "num_tokens": 22059608.0,
      "step": 10630
    },
    {
      "entropy": 5.74952974319458,
      "epoch": 0.9647133526850508,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004914177479536338,
      "loss": 5.6462,
      "mean_token_accuracy": 0.15356021150946617,
      "num_tokens": 22070036.0,
      "step": 10635
    },
    {
      "entropy": 5.7686145305633545,
      "epoch": 0.965166908563135,
      "grad_norm": 2.25,
      "learning_rate": 0.000491408894489034,
      "loss": 5.641,
      "mean_token_accuracy": 0.15436567217111588,
      "num_tokens": 22080679.0,
      "step": 10640
    },
    {
      "entropy": 5.805910730361939,
      "epoch": 0.9656204644412192,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004914000365490432,
      "loss": 5.4927,
      "mean_token_accuracy": 0.15782108306884765,
      "num_tokens": 22092192.0,
      "step": 10645
    },
    {
      "entropy": 5.823235750198364,
      "epoch": 0.9660740203193033,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004913911741338445,
      "loss": 5.6783,
      "mean_token_accuracy": 0.15351950079202653,
      "num_tokens": 22102722.0,
      "step": 10650
    },
    {
      "entropy": 5.76548810005188,
      "epoch": 0.9665275761973875,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004913823072436211,
      "loss": 5.5783,
      "mean_token_accuracy": 0.16051917225122453,
      "num_tokens": 22113521.0,
      "step": 10655
    },
    {
      "entropy": 5.850063276290894,
      "epoch": 0.9669811320754716,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004913734358785565,
      "loss": 5.6703,
      "mean_token_accuracy": 0.15622948259115219,
      "num_tokens": 22124891.0,
      "step": 10660
    },
    {
      "entropy": 5.794833660125732,
      "epoch": 0.9674346879535559,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004913645600388341,
      "loss": 5.6532,
      "mean_token_accuracy": 0.15173319876194,
      "num_tokens": 22135972.0,
      "step": 10665
    },
    {
      "entropy": 5.839829730987549,
      "epoch": 0.9678882438316401,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004913556797246374,
      "loss": 5.7049,
      "mean_token_accuracy": 0.1470730111002922,
      "num_tokens": 22146623.0,
      "step": 10670
    },
    {
      "entropy": 5.847836494445801,
      "epoch": 0.9683417997097242,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004913467949361503,
      "loss": 5.6013,
      "mean_token_accuracy": 0.15360285937786103,
      "num_tokens": 22157619.0,
      "step": 10675
    },
    {
      "entropy": 5.866746807098389,
      "epoch": 0.9687953555878084,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004913379056735561,
      "loss": 5.6142,
      "mean_token_accuracy": 0.1510605663061142,
      "num_tokens": 22168655.0,
      "step": 10680
    },
    {
      "entropy": 5.754996824264526,
      "epoch": 0.9692489114658926,
      "grad_norm": 2.515625,
      "learning_rate": 0.000491329011937039,
      "loss": 5.651,
      "mean_token_accuracy": 0.1540650822222233,
      "num_tokens": 22178500.0,
      "step": 10685
    },
    {
      "entropy": 5.83361701965332,
      "epoch": 0.9697024673439768,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004913201137267828,
      "loss": 5.6513,
      "mean_token_accuracy": 0.1479468896985054,
      "num_tokens": 22188873.0,
      "step": 10690
    },
    {
      "entropy": 5.740324068069458,
      "epoch": 0.970156023222061,
      "grad_norm": 5.125,
      "learning_rate": 0.0004913112110429715,
      "loss": 5.5849,
      "mean_token_accuracy": 0.15538954585790635,
      "num_tokens": 22199743.0,
      "step": 10695
    },
    {
      "entropy": 5.790153455734253,
      "epoch": 0.9706095791001451,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004913023038857892,
      "loss": 5.5389,
      "mean_token_accuracy": 0.16194612979888917,
      "num_tokens": 22210301.0,
      "step": 10700
    },
    {
      "entropy": 5.811965370178223,
      "epoch": 0.9710631349782293,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004912933922554203,
      "loss": 5.547,
      "mean_token_accuracy": 0.16348409205675124,
      "num_tokens": 22220771.0,
      "step": 10705
    },
    {
      "entropy": 5.758093738555909,
      "epoch": 0.9715166908563135,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004912844761520487,
      "loss": 5.6359,
      "mean_token_accuracy": 0.15542892888188362,
      "num_tokens": 22231100.0,
      "step": 10710
    },
    {
      "entropy": 5.736060333251953,
      "epoch": 0.9719702467343977,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004912755555758592,
      "loss": 5.6462,
      "mean_token_accuracy": 0.15574013888835908,
      "num_tokens": 22240997.0,
      "step": 10715
    },
    {
      "entropy": 5.755614709854126,
      "epoch": 0.9724238026124818,
      "grad_norm": 2.6875,
      "learning_rate": 0.000491266630527036,
      "loss": 5.6414,
      "mean_token_accuracy": 0.15689482688903808,
      "num_tokens": 22251795.0,
      "step": 10720
    },
    {
      "entropy": 5.897975873947144,
      "epoch": 0.972877358490566,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004912577010057638,
      "loss": 5.6928,
      "mean_token_accuracy": 0.15143510773777963,
      "num_tokens": 22262465.0,
      "step": 10725
    },
    {
      "entropy": 5.956141090393066,
      "epoch": 0.9733309143686503,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004912487670122273,
      "loss": 5.7718,
      "mean_token_accuracy": 0.15309564918279647,
      "num_tokens": 22272399.0,
      "step": 10730
    },
    {
      "entropy": 5.857673025131225,
      "epoch": 0.9737844702467344,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004912398285466111,
      "loss": 5.6959,
      "mean_token_accuracy": 0.15608857721090316,
      "num_tokens": 22282545.0,
      "step": 10735
    },
    {
      "entropy": 5.825756359100342,
      "epoch": 0.9742380261248186,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004912308856091002,
      "loss": 5.6295,
      "mean_token_accuracy": 0.16046428829431533,
      "num_tokens": 22293154.0,
      "step": 10740
    },
    {
      "entropy": 5.809566068649292,
      "epoch": 0.9746915820029027,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004912219381998795,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15636386275291442,
      "num_tokens": 22302309.0,
      "step": 10745
    },
    {
      "entropy": 5.797397756576538,
      "epoch": 0.9751451378809869,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004912129863191339,
      "loss": 5.6919,
      "mean_token_accuracy": 0.14868070632219316,
      "num_tokens": 22312316.0,
      "step": 10750
    },
    {
      "entropy": 5.913840103149414,
      "epoch": 0.9755986937590712,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004912040299670487,
      "loss": 5.67,
      "mean_token_accuracy": 0.15575771033763885,
      "num_tokens": 22322387.0,
      "step": 10755
    },
    {
      "entropy": 5.8599964618682865,
      "epoch": 0.9760522496371553,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004911950691438091,
      "loss": 5.6383,
      "mean_token_accuracy": 0.16185203492641448,
      "num_tokens": 22332759.0,
      "step": 10760
    },
    {
      "entropy": 5.787653017044067,
      "epoch": 0.9765058055152395,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004911861038496004,
      "loss": 5.6792,
      "mean_token_accuracy": 0.15071214362978935,
      "num_tokens": 22344056.0,
      "step": 10765
    },
    {
      "entropy": 5.744269943237304,
      "epoch": 0.9769593613933236,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004911771340846079,
      "loss": 5.6302,
      "mean_token_accuracy": 0.15291369706392288,
      "num_tokens": 22353877.0,
      "step": 10770
    },
    {
      "entropy": 5.848117399215698,
      "epoch": 0.9774129172714079,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004911681598490172,
      "loss": 5.6493,
      "mean_token_accuracy": 0.1582115352153778,
      "num_tokens": 22363549.0,
      "step": 10775
    },
    {
      "entropy": 5.918741703033447,
      "epoch": 0.977866473149492,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004911591811430139,
      "loss": 5.7284,
      "mean_token_accuracy": 0.15085147470235824,
      "num_tokens": 22374040.0,
      "step": 10780
    },
    {
      "entropy": 5.852246713638306,
      "epoch": 0.9783200290275762,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004911501979667836,
      "loss": 5.7608,
      "mean_token_accuracy": 0.14955372214317322,
      "num_tokens": 22384819.0,
      "step": 10785
    },
    {
      "entropy": 5.825027418136597,
      "epoch": 0.9787735849056604,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004911412103205122,
      "loss": 5.6052,
      "mean_token_accuracy": 0.15450286120176315,
      "num_tokens": 22395130.0,
      "step": 10790
    },
    {
      "entropy": 5.7608222484588625,
      "epoch": 0.9792271407837445,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004911322182043855,
      "loss": 5.6357,
      "mean_token_accuracy": 0.15577088296413422,
      "num_tokens": 22405093.0,
      "step": 10795
    },
    {
      "entropy": 5.802627611160278,
      "epoch": 0.9796806966618288,
      "grad_norm": 2.5,
      "learning_rate": 0.0004911232216185893,
      "loss": 5.6479,
      "mean_token_accuracy": 0.15451014339923858,
      "num_tokens": 22415008.0,
      "step": 10800
    },
    {
      "entropy": 5.888903570175171,
      "epoch": 0.9801342525399129,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004911142205633099,
      "loss": 5.7106,
      "mean_token_accuracy": 0.1523520439863205,
      "num_tokens": 22426486.0,
      "step": 10805
    },
    {
      "entropy": 5.8365579605102536,
      "epoch": 0.9805878084179971,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004911052150387334,
      "loss": 5.6146,
      "mean_token_accuracy": 0.1567578583955765,
      "num_tokens": 22436844.0,
      "step": 10810
    },
    {
      "entropy": 5.847742700576783,
      "epoch": 0.9810413642960812,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004910962050450458,
      "loss": 5.7577,
      "mean_token_accuracy": 0.1515177756547928,
      "num_tokens": 22446630.0,
      "step": 10815
    },
    {
      "entropy": 5.832299661636353,
      "epoch": 0.9814949201741655,
      "grad_norm": 2.375,
      "learning_rate": 0.0004910871905824338,
      "loss": 5.5581,
      "mean_token_accuracy": 0.15815195590257644,
      "num_tokens": 22456406.0,
      "step": 10820
    },
    {
      "entropy": 5.80425329208374,
      "epoch": 0.9819484760522497,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004910781716510837,
      "loss": 5.5762,
      "mean_token_accuracy": 0.16052157580852508,
      "num_tokens": 22466210.0,
      "step": 10825
    },
    {
      "entropy": 5.756770372390747,
      "epoch": 0.9824020319303338,
      "grad_norm": 2.25,
      "learning_rate": 0.0004910691482511817,
      "loss": 5.6691,
      "mean_token_accuracy": 0.1583290547132492,
      "num_tokens": 22476978.0,
      "step": 10830
    },
    {
      "entropy": 5.80738205909729,
      "epoch": 0.982855587808418,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004910601203829148,
      "loss": 5.6339,
      "mean_token_accuracy": 0.15294980108737946,
      "num_tokens": 22488613.0,
      "step": 10835
    },
    {
      "entropy": 5.817235898971558,
      "epoch": 0.9833091436865021,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004910510880464696,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1689889743924141,
      "num_tokens": 22498312.0,
      "step": 10840
    },
    {
      "entropy": 5.792183303833008,
      "epoch": 0.9837626995645864,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004910420512420327,
      "loss": 5.6531,
      "mean_token_accuracy": 0.15933890044689178,
      "num_tokens": 22508131.0,
      "step": 10845
    },
    {
      "entropy": 5.788613033294678,
      "epoch": 0.9842162554426706,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004910330099697913,
      "loss": 5.6496,
      "mean_token_accuracy": 0.15769676193594934,
      "num_tokens": 22517842.0,
      "step": 10850
    },
    {
      "entropy": 5.830097818374634,
      "epoch": 0.9846698113207547,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004910239642299321,
      "loss": 5.62,
      "mean_token_accuracy": 0.15675847083330155,
      "num_tokens": 22528239.0,
      "step": 10855
    },
    {
      "entropy": 5.806100749969483,
      "epoch": 0.9851233671988389,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004910149140226422,
      "loss": 5.5954,
      "mean_token_accuracy": 0.15444757640361786,
      "num_tokens": 22539734.0,
      "step": 10860
    },
    {
      "entropy": 5.815021419525147,
      "epoch": 0.9855769230769231,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004910058593481089,
      "loss": 5.6113,
      "mean_token_accuracy": 0.16095956712961196,
      "num_tokens": 22549796.0,
      "step": 10865
    },
    {
      "entropy": 5.739499568939209,
      "epoch": 0.9860304789550073,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004909968002065195,
      "loss": 5.6115,
      "mean_token_accuracy": 0.15544536262750625,
      "num_tokens": 22559934.0,
      "step": 10870
    },
    {
      "entropy": 5.8420631885528564,
      "epoch": 0.9864840348330914,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000490987736598061,
      "loss": 5.6889,
      "mean_token_accuracy": 0.14803468361496924,
      "num_tokens": 22569864.0,
      "step": 10875
    },
    {
      "entropy": 5.79959135055542,
      "epoch": 0.9869375907111756,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004909786685229211,
      "loss": 5.633,
      "mean_token_accuracy": 0.15979243367910384,
      "num_tokens": 22580687.0,
      "step": 10880
    },
    {
      "entropy": 5.820346117019653,
      "epoch": 0.9873911465892597,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004909695959812875,
      "loss": 5.6595,
      "mean_token_accuracy": 0.15569808185100556,
      "num_tokens": 22590507.0,
      "step": 10885
    },
    {
      "entropy": 5.835328531265259,
      "epoch": 0.987844702467344,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004909605189733474,
      "loss": 5.709,
      "mean_token_accuracy": 0.15876459032297136,
      "num_tokens": 22601901.0,
      "step": 10890
    },
    {
      "entropy": 5.769439935684204,
      "epoch": 0.9882982583454282,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004909514374992889,
      "loss": 5.6411,
      "mean_token_accuracy": 0.153889998793602,
      "num_tokens": 22613003.0,
      "step": 10895
    },
    {
      "entropy": 5.782855749130249,
      "epoch": 0.9887518142235123,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004909423515592996,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16059116572141646,
      "num_tokens": 22623401.0,
      "step": 10900
    },
    {
      "entropy": 5.829384994506836,
      "epoch": 0.9892053701015965,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004909332611535675,
      "loss": 5.6239,
      "mean_token_accuracy": 0.1545998379588127,
      "num_tokens": 22633147.0,
      "step": 10905
    },
    {
      "entropy": 5.774903059005737,
      "epoch": 0.9896589259796807,
      "grad_norm": 2.875,
      "learning_rate": 0.0004909241662822804,
      "loss": 5.492,
      "mean_token_accuracy": 0.1706838309764862,
      "num_tokens": 22642990.0,
      "step": 10910
    },
    {
      "entropy": 5.78953857421875,
      "epoch": 0.9901124818577649,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004909150669456265,
      "loss": 5.6892,
      "mean_token_accuracy": 0.14994772523641586,
      "num_tokens": 22652734.0,
      "step": 10915
    },
    {
      "entropy": 5.848347282409668,
      "epoch": 0.9905660377358491,
      "grad_norm": 3.296875,
      "learning_rate": 0.000490905963143794,
      "loss": 5.7174,
      "mean_token_accuracy": 0.14648480713367462,
      "num_tokens": 22662968.0,
      "step": 10920
    },
    {
      "entropy": 5.813489961624145,
      "epoch": 0.9910195936139332,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004908968548769713,
      "loss": 5.5456,
      "mean_token_accuracy": 0.16475504338741304,
      "num_tokens": 22673157.0,
      "step": 10925
    },
    {
      "entropy": 5.778944778442383,
      "epoch": 0.9914731494920174,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004908877421453466,
      "loss": 5.6798,
      "mean_token_accuracy": 0.15566922426223756,
      "num_tokens": 22683105.0,
      "step": 10930
    },
    {
      "entropy": 5.80912823677063,
      "epoch": 0.9919267053701016,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004908786249491082,
      "loss": 5.6489,
      "mean_token_accuracy": 0.15920983254909515,
      "num_tokens": 22693793.0,
      "step": 10935
    },
    {
      "entropy": 5.945695972442627,
      "epoch": 0.9923802612481858,
      "grad_norm": 2.21875,
      "learning_rate": 0.000490869503288445,
      "loss": 5.6744,
      "mean_token_accuracy": 0.15801735818386078,
      "num_tokens": 22703884.0,
      "step": 10940
    },
    {
      "entropy": 5.860371923446655,
      "epoch": 0.99283381712627,
      "grad_norm": 1.625,
      "learning_rate": 0.0004908603771635454,
      "loss": 5.6948,
      "mean_token_accuracy": 0.15628961175680162,
      "num_tokens": 22713699.0,
      "step": 10945
    },
    {
      "entropy": 5.845581674575806,
      "epoch": 0.9932873730043541,
      "grad_norm": 2.0,
      "learning_rate": 0.0004908512465745982,
      "loss": 5.5425,
      "mean_token_accuracy": 0.16390856802463533,
      "num_tokens": 22722871.0,
      "step": 10950
    },
    {
      "entropy": 5.7688634395599365,
      "epoch": 0.9937409288824384,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004908421115217922,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15266856402158738,
      "num_tokens": 22733314.0,
      "step": 10955
    },
    {
      "entropy": 5.730173349380493,
      "epoch": 0.9941944847605225,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004908329720053164,
      "loss": 5.5687,
      "mean_token_accuracy": 0.1642436295747757,
      "num_tokens": 22743255.0,
      "step": 10960
    },
    {
      "entropy": 5.741264533996582,
      "epoch": 0.9946480406386067,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004908238280253597,
      "loss": 5.561,
      "mean_token_accuracy": 0.16317612677812576,
      "num_tokens": 22753857.0,
      "step": 10965
    },
    {
      "entropy": 5.8027266502380375,
      "epoch": 0.9951015965166908,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004908146795821112,
      "loss": 5.5438,
      "mean_token_accuracy": 0.15576230436563493,
      "num_tokens": 22764326.0,
      "step": 10970
    },
    {
      "entropy": 5.792645835876465,
      "epoch": 0.995555152394775,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004908055266757602,
      "loss": 5.6373,
      "mean_token_accuracy": 0.1557354211807251,
      "num_tokens": 22774958.0,
      "step": 10975
    },
    {
      "entropy": 5.760008525848389,
      "epoch": 0.9960087082728593,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004907963693064958,
      "loss": 5.5343,
      "mean_token_accuracy": 0.15991889238357543,
      "num_tokens": 22785616.0,
      "step": 10980
    },
    {
      "entropy": 5.808274126052856,
      "epoch": 0.9964622641509434,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004907872074745075,
      "loss": 5.7053,
      "mean_token_accuracy": 0.14998844712972642,
      "num_tokens": 22795967.0,
      "step": 10985
    },
    {
      "entropy": 5.737427473068237,
      "epoch": 0.9969158200290276,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004907780411799849,
      "loss": 5.7098,
      "mean_token_accuracy": 0.1508146420121193,
      "num_tokens": 22806335.0,
      "step": 10990
    },
    {
      "entropy": 5.877075958251953,
      "epoch": 0.9973693759071117,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004907688704231173,
      "loss": 5.6294,
      "mean_token_accuracy": 0.1598728522658348,
      "num_tokens": 22816390.0,
      "step": 10995
    },
    {
      "entropy": 5.855796384811401,
      "epoch": 0.997822931785196,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004907596952040945,
      "loss": 5.7025,
      "mean_token_accuracy": 0.15267188623547553,
      "num_tokens": 22826395.0,
      "step": 11000
    },
    {
      "entropy": 5.836028337478638,
      "epoch": 0.9982764876632801,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004907505155231063,
      "loss": 5.685,
      "mean_token_accuracy": 0.1540266364812851,
      "num_tokens": 22836790.0,
      "step": 11005
    },
    {
      "entropy": 5.788539075851441,
      "epoch": 0.9987300435413643,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004907413313803424,
      "loss": 5.5793,
      "mean_token_accuracy": 0.1512584000825882,
      "num_tokens": 22846687.0,
      "step": 11010
    },
    {
      "entropy": 5.874964427947998,
      "epoch": 0.9991835994194485,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004907321427759928,
      "loss": 5.5969,
      "mean_token_accuracy": 0.16132215410470963,
      "num_tokens": 22857208.0,
      "step": 11015
    },
    {
      "entropy": 5.848848915100097,
      "epoch": 0.9996371552975326,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004907229497102475,
      "loss": 5.725,
      "mean_token_accuracy": 0.1487257033586502,
      "num_tokens": 22868556.0,
      "step": 11020
    },
    {
      "entropy": 5.865156602859497,
      "epoch": 1.0000907111756168,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004907137521832966,
      "loss": 5.6894,
      "mean_token_accuracy": 0.15483292788267136,
      "num_tokens": 22880081.0,
      "step": 11025
    },
    {
      "entropy": 5.830396747589111,
      "epoch": 1.000544267053701,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004907045501953304,
      "loss": 5.5247,
      "mean_token_accuracy": 0.16311182975769042,
      "num_tokens": 22892141.0,
      "step": 11030
    },
    {
      "entropy": 5.803026151657105,
      "epoch": 1.0009978229317853,
      "grad_norm": 5.34375,
      "learning_rate": 0.000490695343746539,
      "loss": 5.5888,
      "mean_token_accuracy": 0.16324917674064637,
      "num_tokens": 22902688.0,
      "step": 11035
    },
    {
      "entropy": 5.696486568450927,
      "epoch": 1.0014513788098693,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004906861328371131,
      "loss": 5.4996,
      "mean_token_accuracy": 0.158633553981781,
      "num_tokens": 22912453.0,
      "step": 11040
    },
    {
      "entropy": 5.8159349918365475,
      "epoch": 1.0019049346879536,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004906769174672429,
      "loss": 5.6483,
      "mean_token_accuracy": 0.15360057651996611,
      "num_tokens": 22923167.0,
      "step": 11045
    },
    {
      "entropy": 5.924277257919312,
      "epoch": 1.0023584905660377,
      "grad_norm": 4.53125,
      "learning_rate": 0.000490667697637119,
      "loss": 5.6311,
      "mean_token_accuracy": 0.15385877937078477,
      "num_tokens": 22934753.0,
      "step": 11050
    },
    {
      "entropy": 5.7873869895935055,
      "epoch": 1.002812046444122,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004906584733469324,
      "loss": 5.5552,
      "mean_token_accuracy": 0.16017444133758546,
      "num_tokens": 22945277.0,
      "step": 11055
    },
    {
      "entropy": 5.794849157333374,
      "epoch": 1.0032656023222062,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004906492445968735,
      "loss": 5.6275,
      "mean_token_accuracy": 0.14777041599154472,
      "num_tokens": 22955797.0,
      "step": 11060
    },
    {
      "entropy": 5.851789331436157,
      "epoch": 1.0037191582002902,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004906400113871332,
      "loss": 5.5971,
      "mean_token_accuracy": 0.15662909150123597,
      "num_tokens": 22965969.0,
      "step": 11065
    },
    {
      "entropy": 5.816144752502441,
      "epoch": 1.0041727140783745,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004906307737179026,
      "loss": 5.5523,
      "mean_token_accuracy": 0.1689614936709404,
      "num_tokens": 22975323.0,
      "step": 11070
    },
    {
      "entropy": 5.771370553970337,
      "epoch": 1.0046262699564585,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004906215315893726,
      "loss": 5.5812,
      "mean_token_accuracy": 0.1592397302389145,
      "num_tokens": 22985063.0,
      "step": 11075
    },
    {
      "entropy": 5.851737546920776,
      "epoch": 1.0050798258345428,
      "grad_norm": 4.375,
      "learning_rate": 0.0004906122850017344,
      "loss": 5.6981,
      "mean_token_accuracy": 0.15012871325016022,
      "num_tokens": 22996001.0,
      "step": 11080
    },
    {
      "entropy": 5.851078081130981,
      "epoch": 1.005533381712627,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004906030339551792,
      "loss": 5.6094,
      "mean_token_accuracy": 0.15733970254659652,
      "num_tokens": 23006896.0,
      "step": 11085
    },
    {
      "entropy": 5.830320119857788,
      "epoch": 1.0059869375907111,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004905937784498983,
      "loss": 5.5859,
      "mean_token_accuracy": 0.15726361870765687,
      "num_tokens": 23017080.0,
      "step": 11090
    },
    {
      "entropy": 5.812807703018189,
      "epoch": 1.0064404934687954,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004905845184860831,
      "loss": 5.5377,
      "mean_token_accuracy": 0.16178667098283767,
      "num_tokens": 23027168.0,
      "step": 11095
    },
    {
      "entropy": 5.810239839553833,
      "epoch": 1.0068940493468794,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004905752540639252,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1557539463043213,
      "num_tokens": 23037364.0,
      "step": 11100
    },
    {
      "entropy": 5.766785526275635,
      "epoch": 1.0073476052249637,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004905659851836161,
      "loss": 5.487,
      "mean_token_accuracy": 0.16209156960248947,
      "num_tokens": 23047609.0,
      "step": 11105
    },
    {
      "entropy": 5.8113995552062985,
      "epoch": 1.007801161103048,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004905567118453475,
      "loss": 5.6088,
      "mean_token_accuracy": 0.15477470606565474,
      "num_tokens": 23057942.0,
      "step": 11110
    },
    {
      "entropy": 5.782913875579834,
      "epoch": 1.008254716981132,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004905474340493111,
      "loss": 5.5295,
      "mean_token_accuracy": 0.15781660974025727,
      "num_tokens": 23069119.0,
      "step": 11115
    },
    {
      "entropy": 5.712530899047851,
      "epoch": 1.0087082728592163,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004905381517956988,
      "loss": 5.517,
      "mean_token_accuracy": 0.16303566098213196,
      "num_tokens": 23080060.0,
      "step": 11120
    },
    {
      "entropy": 5.689813899993896,
      "epoch": 1.0091618287373005,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004905288650847027,
      "loss": 5.5316,
      "mean_token_accuracy": 0.15620430260896684,
      "num_tokens": 23090645.0,
      "step": 11125
    },
    {
      "entropy": 5.779735660552978,
      "epoch": 1.0096153846153846,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004905195739165146,
      "loss": 5.5688,
      "mean_token_accuracy": 0.16028851717710496,
      "num_tokens": 23100449.0,
      "step": 11130
    },
    {
      "entropy": 5.780767917633057,
      "epoch": 1.0100689404934688,
      "grad_norm": 3.015625,
      "learning_rate": 0.000490510278291327,
      "loss": 5.5475,
      "mean_token_accuracy": 0.15569900572299958,
      "num_tokens": 23110415.0,
      "step": 11135
    },
    {
      "entropy": 5.787471866607666,
      "epoch": 1.010522496371553,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004905009782093318,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1608660787343979,
      "num_tokens": 23120336.0,
      "step": 11140
    },
    {
      "entropy": 5.753488683700562,
      "epoch": 1.0109760522496372,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004904916736707214,
      "loss": 5.6233,
      "mean_token_accuracy": 0.15254369676113128,
      "num_tokens": 23130568.0,
      "step": 11145
    },
    {
      "entropy": 5.794592237472534,
      "epoch": 1.0114296081277214,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004904823646756885,
      "loss": 5.558,
      "mean_token_accuracy": 0.1587290346622467,
      "num_tokens": 23140080.0,
      "step": 11150
    },
    {
      "entropy": 5.794294214248657,
      "epoch": 1.0118831640058055,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004904730512244252,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1569518581032753,
      "num_tokens": 23150360.0,
      "step": 11155
    },
    {
      "entropy": 5.770283842086792,
      "epoch": 1.0123367198838897,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004904637333171243,
      "loss": 5.5292,
      "mean_token_accuracy": 0.1553722381591797,
      "num_tokens": 23160620.0,
      "step": 11160
    },
    {
      "entropy": 5.8569262504577635,
      "epoch": 1.0127902757619738,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004904544109539785,
      "loss": 5.6203,
      "mean_token_accuracy": 0.15243971943855286,
      "num_tokens": 23171920.0,
      "step": 11165
    },
    {
      "entropy": 5.776814937591553,
      "epoch": 1.013243831640058,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004904450841351805,
      "loss": 5.5694,
      "mean_token_accuracy": 0.1612408071756363,
      "num_tokens": 23182929.0,
      "step": 11170
    },
    {
      "entropy": 5.814702081680298,
      "epoch": 1.0136973875181423,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004904357528609236,
      "loss": 5.5286,
      "mean_token_accuracy": 0.15563077628612518,
      "num_tokens": 23193020.0,
      "step": 11175
    },
    {
      "entropy": 5.7581392288208,
      "epoch": 1.0141509433962264,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004904264171314,
      "loss": 5.5425,
      "mean_token_accuracy": 0.16192737072706223,
      "num_tokens": 23204073.0,
      "step": 11180
    },
    {
      "entropy": 5.703367280960083,
      "epoch": 1.0146044992743106,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004904170769468035,
      "loss": 5.455,
      "mean_token_accuracy": 0.15856247693300246,
      "num_tokens": 23214376.0,
      "step": 11185
    },
    {
      "entropy": 5.829725122451782,
      "epoch": 1.0150580551523947,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004904077323073268,
      "loss": 5.6171,
      "mean_token_accuracy": 0.15695907771587372,
      "num_tokens": 23224304.0,
      "step": 11190
    },
    {
      "entropy": 5.8234271049499515,
      "epoch": 1.015511611030479,
      "grad_norm": 2.75,
      "learning_rate": 0.0004903983832131633,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1551354482769966,
      "num_tokens": 23235324.0,
      "step": 11195
    },
    {
      "entropy": 5.830600929260254,
      "epoch": 1.0159651669085632,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004903890296645063,
      "loss": 5.6004,
      "mean_token_accuracy": 0.15357705280184747,
      "num_tokens": 23245774.0,
      "step": 11200
    },
    {
      "entropy": 5.828565788269043,
      "epoch": 1.0164187227866472,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004903796716615495,
      "loss": 5.544,
      "mean_token_accuracy": 0.15702730417251587,
      "num_tokens": 23256274.0,
      "step": 11205
    },
    {
      "entropy": 5.766544246673584,
      "epoch": 1.0168722786647315,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004903703092044861,
      "loss": 5.5615,
      "mean_token_accuracy": 0.15786381363868712,
      "num_tokens": 23266304.0,
      "step": 11210
    },
    {
      "entropy": 5.700562572479248,
      "epoch": 1.0173258345428158,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004903609422935097,
      "loss": 5.4626,
      "mean_token_accuracy": 0.16196913570165633,
      "num_tokens": 23276552.0,
      "step": 11215
    },
    {
      "entropy": 5.816535758972168,
      "epoch": 1.0177793904208998,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004903515709288143,
      "loss": 5.5879,
      "mean_token_accuracy": 0.1515471652150154,
      "num_tokens": 23287799.0,
      "step": 11220
    },
    {
      "entropy": 5.743448162078858,
      "epoch": 1.018232946298984,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004903421951105933,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16320459693670272,
      "num_tokens": 23297958.0,
      "step": 11225
    },
    {
      "entropy": 5.75546817779541,
      "epoch": 1.0186865021770681,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004903328148390411,
      "loss": 5.6199,
      "mean_token_accuracy": 0.15796598941087722,
      "num_tokens": 23307523.0,
      "step": 11230
    },
    {
      "entropy": 5.872856283187867,
      "epoch": 1.0191400580551524,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004903234301143513,
      "loss": 5.6053,
      "mean_token_accuracy": 0.15784354209899903,
      "num_tokens": 23317970.0,
      "step": 11235
    },
    {
      "entropy": 5.7929973125457765,
      "epoch": 1.0195936139332367,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004903140409367181,
      "loss": 5.5726,
      "mean_token_accuracy": 0.16449209153652192,
      "num_tokens": 23328612.0,
      "step": 11240
    },
    {
      "entropy": 5.77749195098877,
      "epoch": 1.0200471698113207,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004903046473063357,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16105701476335527,
      "num_tokens": 23338104.0,
      "step": 11245
    },
    {
      "entropy": 5.756003093719483,
      "epoch": 1.020500725689405,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004902952492233983,
      "loss": 5.481,
      "mean_token_accuracy": 0.16933009028434753,
      "num_tokens": 23348105.0,
      "step": 11250
    },
    {
      "entropy": 5.69498348236084,
      "epoch": 1.020954281567489,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004902858466881002,
      "loss": 5.4468,
      "mean_token_accuracy": 0.16551368683576584,
      "num_tokens": 23357150.0,
      "step": 11255
    },
    {
      "entropy": 5.802523422241211,
      "epoch": 1.0214078374455733,
      "grad_norm": 3.296875,
      "learning_rate": 0.000490276439700636,
      "loss": 5.6272,
      "mean_token_accuracy": 0.15664812475442885,
      "num_tokens": 23367223.0,
      "step": 11260
    },
    {
      "entropy": 5.800776100158691,
      "epoch": 1.0218613933236576,
      "grad_norm": 2.875,
      "learning_rate": 0.0004902670282612002,
      "loss": 5.5596,
      "mean_token_accuracy": 0.1589964099228382,
      "num_tokens": 23377664.0,
      "step": 11265
    },
    {
      "entropy": 5.713521337509155,
      "epoch": 1.0223149492017416,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004902576123699873,
      "loss": 5.4541,
      "mean_token_accuracy": 0.16042310297489165,
      "num_tokens": 23388057.0,
      "step": 11270
    },
    {
      "entropy": 5.670312547683716,
      "epoch": 1.0227685050798259,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004902481920271922,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16238413229584694,
      "num_tokens": 23397808.0,
      "step": 11275
    },
    {
      "entropy": 5.703090810775757,
      "epoch": 1.02322206095791,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004902387672330097,
      "loss": 5.4935,
      "mean_token_accuracy": 0.1601045623421669,
      "num_tokens": 23406979.0,
      "step": 11280
    },
    {
      "entropy": 5.734795808792114,
      "epoch": 1.0236756168359942,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004902293379876345,
      "loss": 5.519,
      "mean_token_accuracy": 0.16237116158008574,
      "num_tokens": 23416545.0,
      "step": 11285
    },
    {
      "entropy": 5.746735763549805,
      "epoch": 1.0241291727140784,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004902199042912618,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15182949900627135,
      "num_tokens": 23427464.0,
      "step": 11290
    },
    {
      "entropy": 5.789399719238281,
      "epoch": 1.0245827285921625,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004902104661440868,
      "loss": 5.5789,
      "mean_token_accuracy": 0.1616011753678322,
      "num_tokens": 23436690.0,
      "step": 11295
    },
    {
      "entropy": 5.785635852813721,
      "epoch": 1.0250362844702468,
      "grad_norm": 5.75,
      "learning_rate": 0.0004902010235463042,
      "loss": 5.5021,
      "mean_token_accuracy": 0.16081411242485047,
      "num_tokens": 23446701.0,
      "step": 11300
    },
    {
      "entropy": 5.862888479232788,
      "epoch": 1.025489840348331,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004901915764981098,
      "loss": 5.5849,
      "mean_token_accuracy": 0.15415136218070985,
      "num_tokens": 23456966.0,
      "step": 11305
    },
    {
      "entropy": 5.783653259277344,
      "epoch": 1.025943396226415,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004901821249996988,
      "loss": 5.5786,
      "mean_token_accuracy": 0.15850554257631302,
      "num_tokens": 23467300.0,
      "step": 11310
    },
    {
      "entropy": 5.791384220123291,
      "epoch": 1.0263969521044993,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004901726690512665,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15290943831205367,
      "num_tokens": 23478457.0,
      "step": 11315
    },
    {
      "entropy": 5.7736084938049315,
      "epoch": 1.0268505079825834,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004901632086530087,
      "loss": 5.5072,
      "mean_token_accuracy": 0.16404951214790345,
      "num_tokens": 23488066.0,
      "step": 11320
    },
    {
      "entropy": 5.807363986968994,
      "epoch": 1.0273040638606676,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004901537438051207,
      "loss": 5.5865,
      "mean_token_accuracy": 0.15398266762495041,
      "num_tokens": 23498856.0,
      "step": 11325
    },
    {
      "entropy": 5.756262969970703,
      "epoch": 1.027757619738752,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004901442745077986,
      "loss": 5.5447,
      "mean_token_accuracy": 0.15879352241754532,
      "num_tokens": 23509845.0,
      "step": 11330
    },
    {
      "entropy": 5.823694276809692,
      "epoch": 1.028211175616836,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004901348007612381,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1574851855635643,
      "num_tokens": 23520322.0,
      "step": 11335
    },
    {
      "entropy": 5.853392314910889,
      "epoch": 1.0286647314949202,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004901253225656351,
      "loss": 5.5748,
      "mean_token_accuracy": 0.16039203256368637,
      "num_tokens": 23530030.0,
      "step": 11340
    },
    {
      "entropy": 5.866197681427002,
      "epoch": 1.0291182873730043,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004901158399211855,
      "loss": 5.5978,
      "mean_token_accuracy": 0.15863490998744964,
      "num_tokens": 23541063.0,
      "step": 11345
    },
    {
      "entropy": 5.719363069534301,
      "epoch": 1.0295718432510885,
      "grad_norm": 2.125,
      "learning_rate": 0.0004901063528280856,
      "loss": 5.5453,
      "mean_token_accuracy": 0.1542106255888939,
      "num_tokens": 23551693.0,
      "step": 11350
    },
    {
      "entropy": 5.803479194641113,
      "epoch": 1.0300253991291728,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004900968612865314,
      "loss": 5.576,
      "mean_token_accuracy": 0.1636955976486206,
      "num_tokens": 23561758.0,
      "step": 11355
    },
    {
      "entropy": 5.760063648223877,
      "epoch": 1.0304789550072568,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004900873652967195,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1550989605486393,
      "num_tokens": 23571912.0,
      "step": 11360
    },
    {
      "entropy": 5.739124202728272,
      "epoch": 1.030932510885341,
      "grad_norm": 2.75,
      "learning_rate": 0.0004900778648588459,
      "loss": 5.5181,
      "mean_token_accuracy": 0.15988562256097794,
      "num_tokens": 23582600.0,
      "step": 11365
    },
    {
      "entropy": 5.735342407226563,
      "epoch": 1.0313860667634251,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004900683599731073,
      "loss": 5.4133,
      "mean_token_accuracy": 0.1681668356060982,
      "num_tokens": 23592761.0,
      "step": 11370
    },
    {
      "entropy": 5.811415195465088,
      "epoch": 1.0318396226415094,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004900588506397002,
      "loss": 5.6334,
      "mean_token_accuracy": 0.16160874366760253,
      "num_tokens": 23603119.0,
      "step": 11375
    },
    {
      "entropy": 5.729310464859009,
      "epoch": 1.0322931785195937,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004900493368588213,
      "loss": 5.507,
      "mean_token_accuracy": 0.16110349595546722,
      "num_tokens": 23613916.0,
      "step": 11380
    },
    {
      "entropy": 5.793110656738281,
      "epoch": 1.0327467343976777,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004900398186306672,
      "loss": 5.6471,
      "mean_token_accuracy": 0.15355589240789413,
      "num_tokens": 23623956.0,
      "step": 11385
    },
    {
      "entropy": 5.777653598785401,
      "epoch": 1.033200290275762,
      "grad_norm": 2.1875,
      "learning_rate": 0.000490030295955435,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1563473090529442,
      "num_tokens": 23634666.0,
      "step": 11390
    },
    {
      "entropy": 5.7300341606140135,
      "epoch": 1.0336538461538463,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004900207688333215,
      "loss": 5.5063,
      "mean_token_accuracy": 0.15261198133230208,
      "num_tokens": 23645601.0,
      "step": 11395
    },
    {
      "entropy": 5.807222938537597,
      "epoch": 1.0341074020319303,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004900112372645236,
      "loss": 5.5892,
      "mean_token_accuracy": 0.15825920850038527,
      "num_tokens": 23656339.0,
      "step": 11400
    },
    {
      "entropy": 5.798036432266235,
      "epoch": 1.0345609579100146,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004900017012492386,
      "loss": 5.5612,
      "mean_token_accuracy": 0.15829772353172303,
      "num_tokens": 23667433.0,
      "step": 11405
    },
    {
      "entropy": 5.821787405014038,
      "epoch": 1.0350145137880986,
      "grad_norm": 4.875,
      "learning_rate": 0.0004899921607876634,
      "loss": 5.6094,
      "mean_token_accuracy": 0.15375540256500245,
      "num_tokens": 23678186.0,
      "step": 11410
    },
    {
      "entropy": 5.793618154525757,
      "epoch": 1.0354680696661829,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004899826158799958,
      "loss": 5.5524,
      "mean_token_accuracy": 0.1572679951786995,
      "num_tokens": 23688317.0,
      "step": 11415
    },
    {
      "entropy": 5.7853141784667965,
      "epoch": 1.0359216255442671,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004899730665264328,
      "loss": 5.4839,
      "mean_token_accuracy": 0.1556887872517109,
      "num_tokens": 23697752.0,
      "step": 11420
    },
    {
      "entropy": 5.676204586029053,
      "epoch": 1.0363751814223512,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004899635127271721,
      "loss": 5.4778,
      "mean_token_accuracy": 0.16474685668945313,
      "num_tokens": 23707803.0,
      "step": 11425
    },
    {
      "entropy": 5.721381902694702,
      "epoch": 1.0368287373004355,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004899539544824111,
      "loss": 5.4961,
      "mean_token_accuracy": 0.16447029262781143,
      "num_tokens": 23717630.0,
      "step": 11430
    },
    {
      "entropy": 5.858443784713745,
      "epoch": 1.0372822931785195,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004899443917923476,
      "loss": 5.5957,
      "mean_token_accuracy": 0.15795777142047882,
      "num_tokens": 23728836.0,
      "step": 11435
    },
    {
      "entropy": 5.770029544830322,
      "epoch": 1.0377358490566038,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004899348246571792,
      "loss": 5.5849,
      "mean_token_accuracy": 0.15624269396066665,
      "num_tokens": 23740145.0,
      "step": 11440
    },
    {
      "entropy": 5.7938860893249515,
      "epoch": 1.038189404934688,
      "grad_norm": 2.375,
      "learning_rate": 0.000489925253077104,
      "loss": 5.527,
      "mean_token_accuracy": 0.16054621934890748,
      "num_tokens": 23751162.0,
      "step": 11445
    },
    {
      "entropy": 5.730824613571167,
      "epoch": 1.038642960812772,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004899156770523196,
      "loss": 5.5637,
      "mean_token_accuracy": 0.15623183995485307,
      "num_tokens": 23761634.0,
      "step": 11450
    },
    {
      "entropy": 5.724364185333252,
      "epoch": 1.0390965166908563,
      "grad_norm": 2.625,
      "learning_rate": 0.0004899060965830244,
      "loss": 5.5489,
      "mean_token_accuracy": 0.15919629484415054,
      "num_tokens": 23772483.0,
      "step": 11455
    },
    {
      "entropy": 5.785492277145385,
      "epoch": 1.0395500725689404,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004898965116694162,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1587543696165085,
      "num_tokens": 23782298.0,
      "step": 11460
    },
    {
      "entropy": 5.805042123794555,
      "epoch": 1.0400036284470247,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004898869223116935,
      "loss": 5.5577,
      "mean_token_accuracy": 0.15344442874193193,
      "num_tokens": 23792492.0,
      "step": 11465
    },
    {
      "entropy": 5.877156162261963,
      "epoch": 1.040457184325109,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004898773285100546,
      "loss": 5.7103,
      "mean_token_accuracy": 0.1547830045223236,
      "num_tokens": 23802999.0,
      "step": 11470
    },
    {
      "entropy": 5.80629072189331,
      "epoch": 1.040910740203193,
      "grad_norm": 8.5625,
      "learning_rate": 0.0004898677302646975,
      "loss": 5.5281,
      "mean_token_accuracy": 0.16056863516569136,
      "num_tokens": 23812541.0,
      "step": 11475
    },
    {
      "entropy": 5.741022062301636,
      "epoch": 1.0413642960812772,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004898581275758211,
      "loss": 5.6307,
      "mean_token_accuracy": 0.15013584792613982,
      "num_tokens": 23823244.0,
      "step": 11480
    },
    {
      "entropy": 5.7505106925964355,
      "epoch": 1.0418178519593615,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004898485204436239,
      "loss": 5.4879,
      "mean_token_accuracy": 0.16015641987323762,
      "num_tokens": 23832346.0,
      "step": 11485
    },
    {
      "entropy": 5.8216358661651615,
      "epoch": 1.0422714078374455,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004898389088683047,
      "loss": 5.6227,
      "mean_token_accuracy": 0.16134147346019745,
      "num_tokens": 23842554.0,
      "step": 11490
    },
    {
      "entropy": 5.718843460083008,
      "epoch": 1.0427249637155298,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004898292928500619,
      "loss": 5.4529,
      "mean_token_accuracy": 0.16722290962934494,
      "num_tokens": 23852938.0,
      "step": 11495
    },
    {
      "entropy": 5.760653495788574,
      "epoch": 1.0431785195936139,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004898196723890947,
      "loss": 5.4795,
      "mean_token_accuracy": 0.1648701697587967,
      "num_tokens": 23862367.0,
      "step": 11500
    },
    {
      "entropy": 5.74053988456726,
      "epoch": 1.0436320754716981,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004898100474856019,
      "loss": 5.5826,
      "mean_token_accuracy": 0.15636928826570512,
      "num_tokens": 23873249.0,
      "step": 11505
    },
    {
      "entropy": 5.753488063812256,
      "epoch": 1.0440856313497824,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004898004181397824,
      "loss": 5.5644,
      "mean_token_accuracy": 0.15214299708604812,
      "num_tokens": 23883830.0,
      "step": 11510
    },
    {
      "entropy": 5.803683567047119,
      "epoch": 1.0445391872278664,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004897907843518358,
      "loss": 5.5182,
      "mean_token_accuracy": 0.15898152813315392,
      "num_tokens": 23894359.0,
      "step": 11515
    },
    {
      "entropy": 5.731159114837647,
      "epoch": 1.0449927431059507,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004897811461219609,
      "loss": 5.5049,
      "mean_token_accuracy": 0.15634157806634902,
      "num_tokens": 23903779.0,
      "step": 11520
    },
    {
      "entropy": 5.750355339050293,
      "epoch": 1.0454462989840347,
      "grad_norm": 13.5625,
      "learning_rate": 0.0004897715034503571,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15915703177452087,
      "num_tokens": 23914523.0,
      "step": 11525
    },
    {
      "entropy": 5.707683134078979,
      "epoch": 1.045899854862119,
      "grad_norm": 2.375,
      "learning_rate": 0.000489761856337224,
      "loss": 5.5596,
      "mean_token_accuracy": 0.16285839378833772,
      "num_tokens": 23924996.0,
      "step": 11530
    },
    {
      "entropy": 5.734094095230103,
      "epoch": 1.0463534107402033,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004897522047827609,
      "loss": 5.5205,
      "mean_token_accuracy": 0.16182830184698105,
      "num_tokens": 23935843.0,
      "step": 11535
    },
    {
      "entropy": 5.776166391372681,
      "epoch": 1.0468069666182873,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004897425487871676,
      "loss": 5.4958,
      "mean_token_accuracy": 0.1574185460805893,
      "num_tokens": 23945044.0,
      "step": 11540
    },
    {
      "entropy": 5.748601865768433,
      "epoch": 1.0472605224963716,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004897328883506436,
      "loss": 5.5306,
      "mean_token_accuracy": 0.15569346994161606,
      "num_tokens": 23955933.0,
      "step": 11545
    },
    {
      "entropy": 5.775574541091919,
      "epoch": 1.0477140783744556,
      "grad_norm": 2.875,
      "learning_rate": 0.0004897232234733887,
      "loss": 5.5918,
      "mean_token_accuracy": 0.15481627136468887,
      "num_tokens": 23966320.0,
      "step": 11550
    },
    {
      "entropy": 5.746476888656616,
      "epoch": 1.04816763425254,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004897135541556028,
      "loss": 5.5005,
      "mean_token_accuracy": 0.16378279030323029,
      "num_tokens": 23976233.0,
      "step": 11555
    },
    {
      "entropy": 5.771604919433594,
      "epoch": 1.0486211901306242,
      "grad_norm": 2.625,
      "learning_rate": 0.000489703880397486,
      "loss": 5.5711,
      "mean_token_accuracy": 0.15494766235351562,
      "num_tokens": 23987854.0,
      "step": 11560
    },
    {
      "entropy": 5.718685150146484,
      "epoch": 1.0490747460087082,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004896942021992382,
      "loss": 5.562,
      "mean_token_accuracy": 0.15465296059846878,
      "num_tokens": 23997973.0,
      "step": 11565
    },
    {
      "entropy": 5.722238397598266,
      "epoch": 1.0495283018867925,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004896845195610595,
      "loss": 5.5481,
      "mean_token_accuracy": 0.15867221653461455,
      "num_tokens": 24007313.0,
      "step": 11570
    },
    {
      "entropy": 5.768670654296875,
      "epoch": 1.0499818577648767,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004896748324831504,
      "loss": 5.589,
      "mean_token_accuracy": 0.15097645074129104,
      "num_tokens": 24018254.0,
      "step": 11575
    },
    {
      "entropy": 5.832350778579712,
      "epoch": 1.0504354136429608,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004896651409657109,
      "loss": 5.6043,
      "mean_token_accuracy": 0.15535546690225602,
      "num_tokens": 24029799.0,
      "step": 11580
    },
    {
      "entropy": 5.746990394592285,
      "epoch": 1.050888969521045,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004896554450089417,
      "loss": 5.4668,
      "mean_token_accuracy": 0.1698855474591255,
      "num_tokens": 24040972.0,
      "step": 11585
    },
    {
      "entropy": 5.676478624343872,
      "epoch": 1.051342525399129,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004896457446130431,
      "loss": 5.4066,
      "mean_token_accuracy": 0.16618487387895584,
      "num_tokens": 24050781.0,
      "step": 11590
    },
    {
      "entropy": 5.697405385971069,
      "epoch": 1.0517960812772134,
      "grad_norm": 2.5,
      "learning_rate": 0.0004896360397782159,
      "loss": 5.5164,
      "mean_token_accuracy": 0.15654089152812958,
      "num_tokens": 24060184.0,
      "step": 11595
    },
    {
      "entropy": 5.852255058288574,
      "epoch": 1.0522496371552976,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004896263305046607,
      "loss": 5.5832,
      "mean_token_accuracy": 0.1613552138209343,
      "num_tokens": 24070777.0,
      "step": 11600
    },
    {
      "entropy": 5.763235378265381,
      "epoch": 1.0527031930333817,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004896166167925782,
      "loss": 5.5994,
      "mean_token_accuracy": 0.15702067017555238,
      "num_tokens": 24080878.0,
      "step": 11605
    },
    {
      "entropy": 5.739090347290039,
      "epoch": 1.053156748911466,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004896068986421693,
      "loss": 5.5704,
      "mean_token_accuracy": 0.16213036477565765,
      "num_tokens": 24090038.0,
      "step": 11610
    },
    {
      "entropy": 5.837324953079223,
      "epoch": 1.05361030478955,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004895971760536351,
      "loss": 5.603,
      "mean_token_accuracy": 0.16087906658649445,
      "num_tokens": 24100425.0,
      "step": 11615
    },
    {
      "entropy": 5.869068241119384,
      "epoch": 1.0540638606676342,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004895874490271768,
      "loss": 5.6849,
      "mean_token_accuracy": 0.15415374487638472,
      "num_tokens": 24112182.0,
      "step": 11620
    },
    {
      "entropy": 5.788816738128662,
      "epoch": 1.0545174165457185,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004895777175629952,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1549362249672413,
      "num_tokens": 24122182.0,
      "step": 11625
    },
    {
      "entropy": 5.797952890396118,
      "epoch": 1.0549709724238026,
      "grad_norm": 3.625,
      "learning_rate": 0.0004895679816612917,
      "loss": 5.5376,
      "mean_token_accuracy": 0.1583702802658081,
      "num_tokens": 24131930.0,
      "step": 11630
    },
    {
      "entropy": 5.839129638671875,
      "epoch": 1.0554245283018868,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004895582413222678,
      "loss": 5.526,
      "mean_token_accuracy": 0.16291525661945344,
      "num_tokens": 24141575.0,
      "step": 11635
    },
    {
      "entropy": 5.779266023635865,
      "epoch": 1.0558780841799709,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004895484965461247,
      "loss": 5.6657,
      "mean_token_accuracy": 0.155142679810524,
      "num_tokens": 24151720.0,
      "step": 11640
    },
    {
      "entropy": 5.754306221008301,
      "epoch": 1.0563316400580551,
      "grad_norm": 2.5625,
      "learning_rate": 0.000489538747333064,
      "loss": 5.5678,
      "mean_token_accuracy": 0.15946533977985383,
      "num_tokens": 24161881.0,
      "step": 11645
    },
    {
      "entropy": 5.7893461227417,
      "epoch": 1.0567851959361394,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004895289936832874,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1613079935312271,
      "num_tokens": 24171474.0,
      "step": 11650
    },
    {
      "entropy": 5.812449169158936,
      "epoch": 1.0572387518142234,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004895192355969966,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15436174273490905,
      "num_tokens": 24182186.0,
      "step": 11655
    },
    {
      "entropy": 5.797398424148559,
      "epoch": 1.0576923076923077,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004895094730743932,
      "loss": 5.5539,
      "mean_token_accuracy": 0.16231205612421035,
      "num_tokens": 24192318.0,
      "step": 11660
    },
    {
      "entropy": 5.853780698776245,
      "epoch": 1.058145863570392,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004894997061156793,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1608656793832779,
      "num_tokens": 24202918.0,
      "step": 11665
    },
    {
      "entropy": 5.771274995803833,
      "epoch": 1.058599419448476,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004894899347210567,
      "loss": 5.5157,
      "mean_token_accuracy": 0.16073851138353348,
      "num_tokens": 24213218.0,
      "step": 11670
    },
    {
      "entropy": 5.7356054306030275,
      "epoch": 1.0590529753265603,
      "grad_norm": 2.5,
      "learning_rate": 0.0004894801588907278,
      "loss": 5.5161,
      "mean_token_accuracy": 0.16054870188236237,
      "num_tokens": 24223229.0,
      "step": 11675
    },
    {
      "entropy": 5.812273836135864,
      "epoch": 1.0595065312046443,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004894703786248944,
      "loss": 5.581,
      "mean_token_accuracy": 0.15022995471954345,
      "num_tokens": 24233637.0,
      "step": 11680
    },
    {
      "entropy": 5.803047227859497,
      "epoch": 1.0599600870827286,
      "grad_norm": 2.46875,
      "learning_rate": 0.000489460593923759,
      "loss": 5.6094,
      "mean_token_accuracy": 0.16174269765615462,
      "num_tokens": 24244251.0,
      "step": 11685
    },
    {
      "entropy": 5.7527000427246096,
      "epoch": 1.0604136429608129,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004894508047875237,
      "loss": 5.5103,
      "mean_token_accuracy": 0.1596752181649208,
      "num_tokens": 24254380.0,
      "step": 11690
    },
    {
      "entropy": 5.662437391281128,
      "epoch": 1.060867198838897,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004894410112163913,
      "loss": 5.4316,
      "mean_token_accuracy": 0.1607103630900383,
      "num_tokens": 24264841.0,
      "step": 11695
    },
    {
      "entropy": 5.675736808776856,
      "epoch": 1.0613207547169812,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004894312132105641,
      "loss": 5.4988,
      "mean_token_accuracy": 0.16021037101745605,
      "num_tokens": 24274383.0,
      "step": 11700
    },
    {
      "entropy": 5.808772087097168,
      "epoch": 1.0617743105950652,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004894214107702447,
      "loss": 5.5599,
      "mean_token_accuracy": 0.15565950721502303,
      "num_tokens": 24284140.0,
      "step": 11705
    },
    {
      "entropy": 5.893798446655273,
      "epoch": 1.0622278664731495,
      "grad_norm": 2.125,
      "learning_rate": 0.000489411603895636,
      "loss": 5.6526,
      "mean_token_accuracy": 0.1567644879221916,
      "num_tokens": 24295044.0,
      "step": 11710
    },
    {
      "entropy": 5.787691354751587,
      "epoch": 1.0626814223512338,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004894017925869406,
      "loss": 5.5477,
      "mean_token_accuracy": 0.1604714199900627,
      "num_tokens": 24305080.0,
      "step": 11715
    },
    {
      "entropy": 5.7589418411254885,
      "epoch": 1.0631349782293178,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004893919768443615,
      "loss": 5.5929,
      "mean_token_accuracy": 0.15704786777496338,
      "num_tokens": 24316306.0,
      "step": 11720
    },
    {
      "entropy": 5.6547332286834715,
      "epoch": 1.063588534107402,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004893821566681017,
      "loss": 5.5487,
      "mean_token_accuracy": 0.16503504812717437,
      "num_tokens": 24326740.0,
      "step": 11725
    },
    {
      "entropy": 5.787192344665527,
      "epoch": 1.0640420899854863,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004893723320583644,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15400078892707825,
      "num_tokens": 24337621.0,
      "step": 11730
    },
    {
      "entropy": 5.7745507717132565,
      "epoch": 1.0644956458635704,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004893625030153525,
      "loss": 5.4569,
      "mean_token_accuracy": 0.16433921158313752,
      "num_tokens": 24347732.0,
      "step": 11735
    },
    {
      "entropy": 5.791579294204712,
      "epoch": 1.0649492017416546,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004893526695392694,
      "loss": 5.4969,
      "mean_token_accuracy": 0.16680119484663009,
      "num_tokens": 24359173.0,
      "step": 11740
    },
    {
      "entropy": 5.733738374710083,
      "epoch": 1.0654027576197387,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004893428316303185,
      "loss": 5.4729,
      "mean_token_accuracy": 0.1564803808927536,
      "num_tokens": 24369465.0,
      "step": 11745
    },
    {
      "entropy": 5.687932920455933,
      "epoch": 1.065856313497823,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004893329892887031,
      "loss": 5.5126,
      "mean_token_accuracy": 0.1594853550195694,
      "num_tokens": 24378899.0,
      "step": 11750
    },
    {
      "entropy": 5.714597225189209,
      "epoch": 1.0663098693759072,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004893231425146271,
      "loss": 5.6376,
      "mean_token_accuracy": 0.1584230750799179,
      "num_tokens": 24388919.0,
      "step": 11755
    },
    {
      "entropy": 5.748736715316772,
      "epoch": 1.0667634252539913,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004893132913082937,
      "loss": 5.4999,
      "mean_token_accuracy": 0.16259035617113113,
      "num_tokens": 24399632.0,
      "step": 11760
    },
    {
      "entropy": 5.816553211212158,
      "epoch": 1.0672169811320755,
      "grad_norm": 2.5,
      "learning_rate": 0.0004893034356699069,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1614643394947052,
      "num_tokens": 24410817.0,
      "step": 11765
    },
    {
      "entropy": 5.779141235351562,
      "epoch": 1.0676705370101596,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004892935755996705,
      "loss": 5.5481,
      "mean_token_accuracy": 0.15944551825523376,
      "num_tokens": 24421058.0,
      "step": 11770
    },
    {
      "entropy": 5.704815626144409,
      "epoch": 1.0681240928882438,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004892837110977883,
      "loss": 5.4538,
      "mean_token_accuracy": 0.15836425572633744,
      "num_tokens": 24431522.0,
      "step": 11775
    },
    {
      "entropy": 5.716076183319092,
      "epoch": 1.068577648766328,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004892738421644644,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1596313402056694,
      "num_tokens": 24441746.0,
      "step": 11780
    },
    {
      "entropy": 5.7126411437988285,
      "epoch": 1.0690312046444121,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004892639687999027,
      "loss": 5.414,
      "mean_token_accuracy": 0.1638681098818779,
      "num_tokens": 24451916.0,
      "step": 11785
    },
    {
      "entropy": 5.729473066329956,
      "epoch": 1.0694847605224964,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004892540910043076,
      "loss": 5.5454,
      "mean_token_accuracy": 0.1636336326599121,
      "num_tokens": 24461935.0,
      "step": 11790
    },
    {
      "entropy": 5.703885555267334,
      "epoch": 1.0699383164005805,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004892442087778833,
      "loss": 5.4762,
      "mean_token_accuracy": 0.15906477123498916,
      "num_tokens": 24472437.0,
      "step": 11795
    },
    {
      "entropy": 5.71383752822876,
      "epoch": 1.0703918722786647,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004892343221208342,
      "loss": 5.5325,
      "mean_token_accuracy": 0.16436335295438767,
      "num_tokens": 24481726.0,
      "step": 11800
    },
    {
      "entropy": 5.8328804016113285,
      "epoch": 1.070845428156749,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004892244310333646,
      "loss": 5.6385,
      "mean_token_accuracy": 0.14966130405664443,
      "num_tokens": 24492340.0,
      "step": 11805
    },
    {
      "entropy": 5.780572319030762,
      "epoch": 1.071298984034833,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004892145355156793,
      "loss": 5.518,
      "mean_token_accuracy": 0.1623941883444786,
      "num_tokens": 24502138.0,
      "step": 11810
    },
    {
      "entropy": 5.740274572372437,
      "epoch": 1.0717525399129173,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004892046355679829,
      "loss": 5.5688,
      "mean_token_accuracy": 0.15533077716827393,
      "num_tokens": 24512723.0,
      "step": 11815
    },
    {
      "entropy": 5.811892747879028,
      "epoch": 1.0722060957910013,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004891947311904799,
      "loss": 5.6049,
      "mean_token_accuracy": 0.1485086649656296,
      "num_tokens": 24523766.0,
      "step": 11820
    },
    {
      "entropy": 5.812064456939697,
      "epoch": 1.0726596516690856,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004891848223833753,
      "loss": 5.5622,
      "mean_token_accuracy": 0.15951724946498871,
      "num_tokens": 24534571.0,
      "step": 11825
    },
    {
      "entropy": 5.790804481506347,
      "epoch": 1.0731132075471699,
      "grad_norm": 4.1875,
      "learning_rate": 0.000489174909146874,
      "loss": 5.5618,
      "mean_token_accuracy": 0.16514019370079042,
      "num_tokens": 24545265.0,
      "step": 11830
    },
    {
      "entropy": 5.808223009109497,
      "epoch": 1.073566763425254,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004891649914811811,
      "loss": 5.6276,
      "mean_token_accuracy": 0.1587499588727951,
      "num_tokens": 24555524.0,
      "step": 11835
    },
    {
      "entropy": 5.773516654968262,
      "epoch": 1.0740203193033382,
      "grad_norm": 7.75,
      "learning_rate": 0.0004891550693865014,
      "loss": 5.5005,
      "mean_token_accuracy": 0.1684954434633255,
      "num_tokens": 24565420.0,
      "step": 11840
    },
    {
      "entropy": 5.808403062820434,
      "epoch": 1.0744738751814222,
      "grad_norm": 3.875,
      "learning_rate": 0.0004891451428630405,
      "loss": 5.5525,
      "mean_token_accuracy": 0.15781337320804595,
      "num_tokens": 24576794.0,
      "step": 11845
    },
    {
      "entropy": 5.762145185470581,
      "epoch": 1.0749274310595065,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004891352119110033,
      "loss": 5.5556,
      "mean_token_accuracy": 0.16086098402738572,
      "num_tokens": 24586320.0,
      "step": 11850
    },
    {
      "entropy": 5.7493363380432125,
      "epoch": 1.0753809869375908,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004891252765305954,
      "loss": 5.5305,
      "mean_token_accuracy": 0.16228426843881608,
      "num_tokens": 24596524.0,
      "step": 11855
    },
    {
      "entropy": 5.799742603302002,
      "epoch": 1.0758345428156748,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004891153367220222,
      "loss": 5.4769,
      "mean_token_accuracy": 0.16120700389146805,
      "num_tokens": 24605877.0,
      "step": 11860
    },
    {
      "entropy": 5.871392726898193,
      "epoch": 1.076288098693759,
      "grad_norm": 2.625,
      "learning_rate": 0.0004891053924854894,
      "loss": 5.6675,
      "mean_token_accuracy": 0.15001487284898757,
      "num_tokens": 24616719.0,
      "step": 11865
    },
    {
      "entropy": 5.732544708251953,
      "epoch": 1.0767416545718433,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004890954438212024,
      "loss": 5.4695,
      "mean_token_accuracy": 0.16442711055278778,
      "num_tokens": 24626882.0,
      "step": 11870
    },
    {
      "entropy": 5.762554931640625,
      "epoch": 1.0771952104499274,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004890854907293671,
      "loss": 5.5507,
      "mean_token_accuracy": 0.15483054518699646,
      "num_tokens": 24637884.0,
      "step": 11875
    },
    {
      "entropy": 5.720759391784668,
      "epoch": 1.0776487663280117,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004890755332101893,
      "loss": 5.5107,
      "mean_token_accuracy": 0.15827159583568573,
      "num_tokens": 24647389.0,
      "step": 11880
    },
    {
      "entropy": 5.756432104110718,
      "epoch": 1.0781023222060957,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004890655712638749,
      "loss": 5.5659,
      "mean_token_accuracy": 0.15237744450569152,
      "num_tokens": 24657595.0,
      "step": 11885
    },
    {
      "entropy": 5.75420880317688,
      "epoch": 1.07855587808418,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048905560489063,
      "loss": 5.5033,
      "mean_token_accuracy": 0.15938466638326645,
      "num_tokens": 24667683.0,
      "step": 11890
    },
    {
      "entropy": 5.859339427947998,
      "epoch": 1.0790094339622642,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004890456340906605,
      "loss": 5.6284,
      "mean_token_accuracy": 0.1514022320508957,
      "num_tokens": 24678041.0,
      "step": 11895
    },
    {
      "entropy": 5.78812313079834,
      "epoch": 1.0794629898403483,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004890356588641729,
      "loss": 5.6002,
      "mean_token_accuracy": 0.16286241114139557,
      "num_tokens": 24688569.0,
      "step": 11900
    },
    {
      "entropy": 5.7869555950164795,
      "epoch": 1.0799165457184325,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004890256792113733,
      "loss": 5.5975,
      "mean_token_accuracy": 0.15714042633771896,
      "num_tokens": 24699913.0,
      "step": 11905
    },
    {
      "entropy": 5.867937850952148,
      "epoch": 1.0803701015965168,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004890156951324681,
      "loss": 5.588,
      "mean_token_accuracy": 0.15711037516593934,
      "num_tokens": 24710682.0,
      "step": 11910
    },
    {
      "entropy": 5.890251636505127,
      "epoch": 1.0808236574746009,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004890057066276636,
      "loss": 5.678,
      "mean_token_accuracy": 0.15511685758829116,
      "num_tokens": 24721087.0,
      "step": 11915
    },
    {
      "entropy": 5.749015665054321,
      "epoch": 1.0812772133526851,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004889957136971668,
      "loss": 5.5046,
      "mean_token_accuracy": 0.16428616046905517,
      "num_tokens": 24731323.0,
      "step": 11920
    },
    {
      "entropy": 5.734285354614258,
      "epoch": 1.0817307692307692,
      "grad_norm": 2.390625,
      "learning_rate": 0.000488985716341184,
      "loss": 5.5549,
      "mean_token_accuracy": 0.15688968300819398,
      "num_tokens": 24742559.0,
      "step": 11925
    },
    {
      "entropy": 5.8322375297546385,
      "epoch": 1.0821843251088534,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004889757145599222,
      "loss": 5.5385,
      "mean_token_accuracy": 0.158398574590683,
      "num_tokens": 24752307.0,
      "step": 11930
    },
    {
      "entropy": 5.715606069564819,
      "epoch": 1.0826378809869377,
      "grad_norm": 3.40625,
      "learning_rate": 0.000488965708353588,
      "loss": 5.4942,
      "mean_token_accuracy": 0.15804069489240646,
      "num_tokens": 24762792.0,
      "step": 11935
    },
    {
      "entropy": 5.799453639984131,
      "epoch": 1.0830914368650217,
      "grad_norm": 3.75,
      "learning_rate": 0.0004889556977223883,
      "loss": 5.6277,
      "mean_token_accuracy": 0.1521786704659462,
      "num_tokens": 24774377.0,
      "step": 11940
    },
    {
      "entropy": 5.814619016647339,
      "epoch": 1.083544992743106,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004889456826665304,
      "loss": 5.5527,
      "mean_token_accuracy": 0.15799938887357712,
      "num_tokens": 24785919.0,
      "step": 11945
    },
    {
      "entropy": 5.841064405441284,
      "epoch": 1.08399854862119,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004889356631862211,
      "loss": 5.5847,
      "mean_token_accuracy": 0.15743543207645416,
      "num_tokens": 24796594.0,
      "step": 11950
    },
    {
      "entropy": 5.763912773132324,
      "epoch": 1.0844521044992743,
      "grad_norm": 2.75,
      "learning_rate": 0.000488925639281668,
      "loss": 5.565,
      "mean_token_accuracy": 0.15561186671257018,
      "num_tokens": 24807322.0,
      "step": 11955
    },
    {
      "entropy": 5.741538190841675,
      "epoch": 1.0849056603773586,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004889156109530781,
      "loss": 5.4712,
      "mean_token_accuracy": 0.16155950129032134,
      "num_tokens": 24817593.0,
      "step": 11960
    },
    {
      "entropy": 5.769239807128907,
      "epoch": 1.0853592162554426,
      "grad_norm": 3.5,
      "learning_rate": 0.0004889055782006588,
      "loss": 5.4859,
      "mean_token_accuracy": 0.15776087194681168,
      "num_tokens": 24827577.0,
      "step": 11965
    },
    {
      "entropy": 5.742091464996338,
      "epoch": 1.085812772133527,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004888955410246176,
      "loss": 5.5279,
      "mean_token_accuracy": 0.1618715599179268,
      "num_tokens": 24838148.0,
      "step": 11970
    },
    {
      "entropy": 5.732159185409546,
      "epoch": 1.086266328011611,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004888854994251621,
      "loss": 5.4838,
      "mean_token_accuracy": 0.16252688318490982,
      "num_tokens": 24848451.0,
      "step": 11975
    },
    {
      "entropy": 5.761207103729248,
      "epoch": 1.0867198838896952,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004888754534025,
      "loss": 5.5747,
      "mean_token_accuracy": 0.15174014270305633,
      "num_tokens": 24859619.0,
      "step": 11980
    },
    {
      "entropy": 5.761291885375977,
      "epoch": 1.0871734397677795,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004888654029568391,
      "loss": 5.6151,
      "mean_token_accuracy": 0.14883196651935576,
      "num_tokens": 24869863.0,
      "step": 11985
    },
    {
      "entropy": 5.742021799087524,
      "epoch": 1.0876269956458635,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004888553480883871,
      "loss": 5.5408,
      "mean_token_accuracy": 0.15269162356853486,
      "num_tokens": 24880524.0,
      "step": 11990
    },
    {
      "entropy": 5.733129167556763,
      "epoch": 1.0880805515239478,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000488845288797352,
      "loss": 5.5186,
      "mean_token_accuracy": 0.16384661942720413,
      "num_tokens": 24891514.0,
      "step": 11995
    },
    {
      "entropy": 5.739286851882935,
      "epoch": 1.0885341074020318,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004888352250839418,
      "loss": 5.5016,
      "mean_token_accuracy": 0.16507037729024887,
      "num_tokens": 24900842.0,
      "step": 12000
    },
    {
      "epoch": 1.0885341074020318,
      "eval_entropy": 5.580312384091891,
      "eval_loss": 5.6294732093811035,
      "eval_mean_token_accuracy": 0.16381040554982504,
      "eval_num_tokens": 24900842.0,
      "eval_runtime": 26.6163,
      "eval_samples_per_second": 1328.511,
      "eval_steps_per_second": 166.064,
      "step": 12000
    },
    {
      "entropy": 5.770009183883667,
      "epoch": 1.088987663280116,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004888251569483647,
      "loss": 5.4748,
      "mean_token_accuracy": 0.16381522864103318,
      "num_tokens": 24911845.0,
      "step": 12005
    },
    {
      "entropy": 5.69474573135376,
      "epoch": 1.0894412191582004,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004888150843908288,
      "loss": 5.5057,
      "mean_token_accuracy": 0.16607357412576676,
      "num_tokens": 24922211.0,
      "step": 12010
    },
    {
      "entropy": 5.809834051132202,
      "epoch": 1.0898947750362844,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004888050074115426,
      "loss": 5.6049,
      "mean_token_accuracy": 0.15531208217144013,
      "num_tokens": 24932129.0,
      "step": 12015
    },
    {
      "entropy": 5.897721672058106,
      "epoch": 1.0903483309143687,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004887949260107143,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1558338910341263,
      "num_tokens": 24942869.0,
      "step": 12020
    },
    {
      "entropy": 5.700602102279663,
      "epoch": 1.0908018867924527,
      "grad_norm": 2.375,
      "learning_rate": 0.0004887848401885524,
      "loss": 5.5418,
      "mean_token_accuracy": 0.16099727153778076,
      "num_tokens": 24952621.0,
      "step": 12025
    },
    {
      "entropy": 5.804478645324707,
      "epoch": 1.091255442670537,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004887747499452656,
      "loss": 5.6422,
      "mean_token_accuracy": 0.14877880066633226,
      "num_tokens": 24963587.0,
      "step": 12030
    },
    {
      "entropy": 5.812208986282348,
      "epoch": 1.0917089985486212,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004887646552810624,
      "loss": 5.5222,
      "mean_token_accuracy": 0.16232824176549912,
      "num_tokens": 24974122.0,
      "step": 12035
    },
    {
      "entropy": 5.769900608062744,
      "epoch": 1.0921625544267053,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004887545561961517,
      "loss": 5.5749,
      "mean_token_accuracy": 0.1595664232969284,
      "num_tokens": 24985145.0,
      "step": 12040
    },
    {
      "entropy": 5.682806444168091,
      "epoch": 1.0926161103047896,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004887444526907424,
      "loss": 5.5134,
      "mean_token_accuracy": 0.1577943593263626,
      "num_tokens": 24995511.0,
      "step": 12045
    },
    {
      "entropy": 5.788463973999024,
      "epoch": 1.0930696661828738,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004887343447650432,
      "loss": 5.5433,
      "mean_token_accuracy": 0.15686437487602234,
      "num_tokens": 25006193.0,
      "step": 12050
    },
    {
      "entropy": 5.835666179656982,
      "epoch": 1.0935232220609579,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004887242324192634,
      "loss": 5.5776,
      "mean_token_accuracy": 0.15770234316587448,
      "num_tokens": 25016444.0,
      "step": 12055
    },
    {
      "entropy": 5.814560747146606,
      "epoch": 1.0939767779390421,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004887141156536119,
      "loss": 5.6005,
      "mean_token_accuracy": 0.15352796018123627,
      "num_tokens": 25027898.0,
      "step": 12060
    },
    {
      "entropy": 5.7406800270080565,
      "epoch": 1.0944303338171262,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004887039944682981,
      "loss": 5.4594,
      "mean_token_accuracy": 0.15977587252855302,
      "num_tokens": 25038201.0,
      "step": 12065
    },
    {
      "entropy": 5.869744539260864,
      "epoch": 1.0948838896952104,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004886938688635313,
      "loss": 5.7049,
      "mean_token_accuracy": 0.14768646210432052,
      "num_tokens": 25048466.0,
      "step": 12070
    },
    {
      "entropy": 5.886350059509278,
      "epoch": 1.0953374455732947,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004886837388395207,
      "loss": 5.6184,
      "mean_token_accuracy": 0.15671121329069138,
      "num_tokens": 25059064.0,
      "step": 12075
    },
    {
      "entropy": 5.817844581604004,
      "epoch": 1.0957910014513788,
      "grad_norm": 13.6875,
      "learning_rate": 0.000488673604396476,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1522151604294777,
      "num_tokens": 25068834.0,
      "step": 12080
    },
    {
      "entropy": 5.864861536026001,
      "epoch": 1.096244557329463,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004886634655346067,
      "loss": 5.6257,
      "mean_token_accuracy": 0.14941567406058312,
      "num_tokens": 25080058.0,
      "step": 12085
    },
    {
      "entropy": 5.785624217987061,
      "epoch": 1.0966981132075473,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004886533222541225,
      "loss": 5.5117,
      "mean_token_accuracy": 0.1651437073945999,
      "num_tokens": 25089751.0,
      "step": 12090
    },
    {
      "entropy": 5.765842628479004,
      "epoch": 1.0971516690856313,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004886431745552331,
      "loss": 5.5568,
      "mean_token_accuracy": 0.1582575961947441,
      "num_tokens": 25099055.0,
      "step": 12095
    },
    {
      "entropy": 5.8358715057373045,
      "epoch": 1.0976052249637156,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004886330224381485,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1507570207118988,
      "num_tokens": 25109793.0,
      "step": 12100
    },
    {
      "entropy": 5.828896617889404,
      "epoch": 1.0980587808417996,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004886228659030785,
      "loss": 5.5898,
      "mean_token_accuracy": 0.16082045584917068,
      "num_tokens": 25119814.0,
      "step": 12105
    },
    {
      "entropy": 5.836939334869385,
      "epoch": 1.098512336719884,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004886127049502331,
      "loss": 5.6032,
      "mean_token_accuracy": 0.15646470710635185,
      "num_tokens": 25130747.0,
      "step": 12110
    },
    {
      "entropy": 5.752117490768432,
      "epoch": 1.0989658925979682,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004886025395798228,
      "loss": 5.5337,
      "mean_token_accuracy": 0.15877415984869003,
      "num_tokens": 25142235.0,
      "step": 12115
    },
    {
      "entropy": 5.665432596206665,
      "epoch": 1.0994194484760522,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004885923697920573,
      "loss": 5.4537,
      "mean_token_accuracy": 0.164444237947464,
      "num_tokens": 25152047.0,
      "step": 12120
    },
    {
      "entropy": 5.717042255401611,
      "epoch": 1.0998730043541365,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004885821955871472,
      "loss": 5.5184,
      "mean_token_accuracy": 0.1643191769719124,
      "num_tokens": 25162678.0,
      "step": 12125
    },
    {
      "entropy": 5.803634405136108,
      "epoch": 1.1003265602322205,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004885720169653028,
      "loss": 5.5126,
      "mean_token_accuracy": 0.16183817982673646,
      "num_tokens": 25172973.0,
      "step": 12130
    },
    {
      "entropy": 5.824831199645996,
      "epoch": 1.1007801161103048,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004885618339267348,
      "loss": 5.61,
      "mean_token_accuracy": 0.15516517907381058,
      "num_tokens": 25183458.0,
      "step": 12135
    },
    {
      "entropy": 5.767845010757446,
      "epoch": 1.101233671988389,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004885516464716537,
      "loss": 5.5544,
      "mean_token_accuracy": 0.15890781730413436,
      "num_tokens": 25193827.0,
      "step": 12140
    },
    {
      "entropy": 5.8302689552307125,
      "epoch": 1.101687227866473,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048854145460027,
      "loss": 5.5567,
      "mean_token_accuracy": 0.1549244061112404,
      "num_tokens": 25204448.0,
      "step": 12145
    },
    {
      "entropy": 5.807820844650268,
      "epoch": 1.1021407837445574,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004885312583127946,
      "loss": 5.5057,
      "mean_token_accuracy": 0.16087655127048492,
      "num_tokens": 25215041.0,
      "step": 12150
    },
    {
      "entropy": 5.750949048995972,
      "epoch": 1.1025943396226414,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004885210576094385,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1563236653804779,
      "num_tokens": 25225937.0,
      "step": 12155
    },
    {
      "entropy": 5.7224174499511715,
      "epoch": 1.1030478955007257,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004885108524904124,
      "loss": 5.4615,
      "mean_token_accuracy": 0.16081289798021317,
      "num_tokens": 25235270.0,
      "step": 12160
    },
    {
      "entropy": 5.7585631847381595,
      "epoch": 1.10350145137881,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004885006429559275,
      "loss": 5.5758,
      "mean_token_accuracy": 0.1516699232161045,
      "num_tokens": 25247080.0,
      "step": 12165
    },
    {
      "entropy": 5.829455804824829,
      "epoch": 1.103955007256894,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004884904290061949,
      "loss": 5.5368,
      "mean_token_accuracy": 0.1535037413239479,
      "num_tokens": 25256718.0,
      "step": 12170
    },
    {
      "entropy": 5.824248218536377,
      "epoch": 1.1044085631349783,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004884802106414258,
      "loss": 5.6481,
      "mean_token_accuracy": 0.15341632515192033,
      "num_tokens": 25266632.0,
      "step": 12175
    },
    {
      "entropy": 5.7874068260192875,
      "epoch": 1.1048621190130623,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004884699878618316,
      "loss": 5.5104,
      "mean_token_accuracy": 0.16062667965888977,
      "num_tokens": 25275598.0,
      "step": 12180
    },
    {
      "entropy": 5.806826066970825,
      "epoch": 1.1053156748911466,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004884597606676236,
      "loss": 5.6333,
      "mean_token_accuracy": 0.1572799041867256,
      "num_tokens": 25286598.0,
      "step": 12185
    },
    {
      "entropy": 5.770541667938232,
      "epoch": 1.1057692307692308,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004884495290590134,
      "loss": 5.5261,
      "mean_token_accuracy": 0.1630978688597679,
      "num_tokens": 25296730.0,
      "step": 12190
    },
    {
      "entropy": 5.728164386749268,
      "epoch": 1.1062227866473149,
      "grad_norm": 2.875,
      "learning_rate": 0.0004884392930362124,
      "loss": 5.4551,
      "mean_token_accuracy": 0.16584885418415068,
      "num_tokens": 25306326.0,
      "step": 12195
    },
    {
      "entropy": 5.738533973693848,
      "epoch": 1.1066763425253991,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004884290525994326,
      "loss": 5.5737,
      "mean_token_accuracy": 0.1590038001537323,
      "num_tokens": 25316342.0,
      "step": 12200
    },
    {
      "entropy": 5.827234601974487,
      "epoch": 1.1071298984034832,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004884188077488855,
      "loss": 5.6083,
      "mean_token_accuracy": 0.15078015252947807,
      "num_tokens": 25326129.0,
      "step": 12205
    },
    {
      "entropy": 5.798996543884277,
      "epoch": 1.1075834542815675,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004884085584847831,
      "loss": 5.5781,
      "mean_token_accuracy": 0.15619731843471527,
      "num_tokens": 25337156.0,
      "step": 12210
    },
    {
      "entropy": 5.8891950130462645,
      "epoch": 1.1080370101596517,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004883983048073375,
      "loss": 5.6691,
      "mean_token_accuracy": 0.15022148191928864,
      "num_tokens": 25347632.0,
      "step": 12215
    },
    {
      "entropy": 5.768394136428833,
      "epoch": 1.1084905660377358,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004883880467167603,
      "loss": 5.4617,
      "mean_token_accuracy": 0.1601735919713974,
      "num_tokens": 25359186.0,
      "step": 12220
    },
    {
      "entropy": 5.77440710067749,
      "epoch": 1.10894412191582,
      "grad_norm": 2.625,
      "learning_rate": 0.000488377784213264,
      "loss": 5.5649,
      "mean_token_accuracy": 0.16213472038507462,
      "num_tokens": 25368993.0,
      "step": 12225
    },
    {
      "entropy": 5.743676805496216,
      "epoch": 1.1093976777939043,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004883675172970608,
      "loss": 5.518,
      "mean_token_accuracy": 0.16184825599193572,
      "num_tokens": 25379880.0,
      "step": 12230
    },
    {
      "entropy": 5.771228981018067,
      "epoch": 1.1098512336719883,
      "grad_norm": 2.3125,
      "learning_rate": 0.000488357245968363,
      "loss": 5.707,
      "mean_token_accuracy": 0.15197538435459138,
      "num_tokens": 25389942.0,
      "step": 12235
    },
    {
      "entropy": 5.792243576049804,
      "epoch": 1.1103047895500726,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004883469702273829,
      "loss": 5.5051,
      "mean_token_accuracy": 0.16346201449632644,
      "num_tokens": 25400836.0,
      "step": 12240
    },
    {
      "entropy": 5.7247536182403564,
      "epoch": 1.1107583454281567,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004883366900743331,
      "loss": 5.5314,
      "mean_token_accuracy": 0.16693134605884552,
      "num_tokens": 25410502.0,
      "step": 12245
    },
    {
      "entropy": 5.732777214050293,
      "epoch": 1.111211901306241,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004883264055094262,
      "loss": 5.4675,
      "mean_token_accuracy": 0.1655327558517456,
      "num_tokens": 25420301.0,
      "step": 12250
    },
    {
      "entropy": 5.780748271942139,
      "epoch": 1.1116654571843252,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004883161165328749,
      "loss": 5.552,
      "mean_token_accuracy": 0.16022371798753737,
      "num_tokens": 25431107.0,
      "step": 12255
    },
    {
      "entropy": 5.783886289596557,
      "epoch": 1.1121190130624092,
      "grad_norm": 3.390625,
      "learning_rate": 0.000488305823144892,
      "loss": 5.5372,
      "mean_token_accuracy": 0.1599503993988037,
      "num_tokens": 25441256.0,
      "step": 12260
    },
    {
      "entropy": 5.805974388122559,
      "epoch": 1.1125725689404935,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004882955253456902,
      "loss": 5.5013,
      "mean_token_accuracy": 0.16793554723262788,
      "num_tokens": 25451627.0,
      "step": 12265
    },
    {
      "entropy": 5.783116960525513,
      "epoch": 1.1130261248185778,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004882852231354826,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15016109123826027,
      "num_tokens": 25461273.0,
      "step": 12270
    },
    {
      "entropy": 5.756811857223511,
      "epoch": 1.1134796806966618,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004882749165144822,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16313402801752092,
      "num_tokens": 25471949.0,
      "step": 12275
    },
    {
      "entropy": 5.831482982635498,
      "epoch": 1.113933236574746,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004882646054829023,
      "loss": 5.529,
      "mean_token_accuracy": 0.15861806422472,
      "num_tokens": 25482175.0,
      "step": 12280
    },
    {
      "entropy": 5.76349778175354,
      "epoch": 1.1143867924528301,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048825429004095597,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1565583363175392,
      "num_tokens": 25491786.0,
      "step": 12285
    },
    {
      "entropy": 5.748883438110352,
      "epoch": 1.1148403483309144,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004882439701888566,
      "loss": 5.5234,
      "mean_token_accuracy": 0.1675867274403572,
      "num_tokens": 25502103.0,
      "step": 12290
    },
    {
      "entropy": 5.740812635421753,
      "epoch": 1.1152939042089987,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048823364592681754,
      "loss": 5.4765,
      "mean_token_accuracy": 0.16189854294061662,
      "num_tokens": 25512530.0,
      "step": 12295
    },
    {
      "entropy": 5.7487006187438965,
      "epoch": 1.1157474600870827,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048822331725505234,
      "loss": 5.4954,
      "mean_token_accuracy": 0.16113103926181793,
      "num_tokens": 25522905.0,
      "step": 12300
    },
    {
      "entropy": 5.758361911773681,
      "epoch": 1.116201015965167,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048821298417377456,
      "loss": 5.4804,
      "mean_token_accuracy": 0.1642244726419449,
      "num_tokens": 25532963.0,
      "step": 12305
    },
    {
      "entropy": 5.664602994918823,
      "epoch": 1.116654571843251,
      "grad_norm": 2.390625,
      "learning_rate": 0.000488202646683198,
      "loss": 5.5469,
      "mean_token_accuracy": 0.1564118593931198,
      "num_tokens": 25542583.0,
      "step": 12310
    },
    {
      "entropy": 5.701724052429199,
      "epoch": 1.1171081277213353,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004881923047835364,
      "loss": 5.5175,
      "mean_token_accuracy": 0.16303540021181107,
      "num_tokens": 25553367.0,
      "step": 12315
    },
    {
      "entropy": 5.925316429138183,
      "epoch": 1.1175616835994195,
      "grad_norm": 5.6875,
      "learning_rate": 0.0004881819584750036,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15699404031038283,
      "num_tokens": 25564823.0,
      "step": 12320
    },
    {
      "entropy": 5.821801376342774,
      "epoch": 1.1180152394775036,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048817160775781356,
      "loss": 5.5776,
      "mean_token_accuracy": 0.16198044866323472,
      "num_tokens": 25574946.0,
      "step": 12325
    },
    {
      "entropy": 5.765165090560913,
      "epoch": 1.1184687953555879,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048816125263218023,
      "loss": 5.5677,
      "mean_token_accuracy": 0.1549558162689209,
      "num_tokens": 25584779.0,
      "step": 12330
    },
    {
      "entropy": 5.80529088973999,
      "epoch": 1.118922351233672,
      "grad_norm": 3.0,
      "learning_rate": 0.00048815089309831803,
      "loss": 5.6294,
      "mean_token_accuracy": 0.15640642791986464,
      "num_tokens": 25594725.0,
      "step": 12335
    },
    {
      "entropy": 5.817852544784546,
      "epoch": 1.1193759071117562,
      "grad_norm": 2.171875,
      "learning_rate": 0.000488140529156441,
      "loss": 5.6291,
      "mean_token_accuracy": 0.15307459235191345,
      "num_tokens": 25605499.0,
      "step": 12340
    },
    {
      "entropy": 5.7613097667694095,
      "epoch": 1.1198294629898404,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004881301608067635,
      "loss": 5.5537,
      "mean_token_accuracy": 0.1632678210735321,
      "num_tokens": 25616443.0,
      "step": 12345
    },
    {
      "entropy": 5.796557188034058,
      "epoch": 1.1202830188679245,
      "grad_norm": 4.375,
      "learning_rate": 0.00048811978804949997,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1544956810772419,
      "num_tokens": 25627762.0,
      "step": 12350
    },
    {
      "entropy": 5.830805349349975,
      "epoch": 1.1207365747460087,
      "grad_norm": 2.703125,
      "learning_rate": 0.000488109410884865,
      "loss": 5.6103,
      "mean_token_accuracy": 0.15734838992357253,
      "num_tokens": 25637904.0,
      "step": 12355
    },
    {
      "entropy": 5.7924089431762695,
      "epoch": 1.1211901306240928,
      "grad_norm": 4.1875,
      "learning_rate": 0.00048809902931307294,
      "loss": 5.4898,
      "mean_token_accuracy": 0.15840931385755538,
      "num_tokens": 25649058.0,
      "step": 12360
    },
    {
      "entropy": 5.7830061435699465,
      "epoch": 1.121643686502177,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004880886433343388,
      "loss": 5.5586,
      "mean_token_accuracy": 0.16376760900020598,
      "num_tokens": 25659161.0,
      "step": 12365
    },
    {
      "entropy": 5.7729823112487795,
      "epoch": 1.1220972423802613,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004880782529488771,
      "loss": 5.4779,
      "mean_token_accuracy": 0.16279046833515168,
      "num_tokens": 25669336.0,
      "step": 12370
    },
    {
      "entropy": 5.790310525894165,
      "epoch": 1.1225507982583454,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004880678581569029,
      "loss": 5.6086,
      "mean_token_accuracy": 0.1595725819468498,
      "num_tokens": 25680443.0,
      "step": 12375
    },
    {
      "entropy": 5.766762542724609,
      "epoch": 1.1230043541364296,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048805745895863105,
      "loss": 5.5048,
      "mean_token_accuracy": 0.1644008204340935,
      "num_tokens": 25690202.0,
      "step": 12380
    },
    {
      "entropy": 5.7905182361602785,
      "epoch": 1.1234579100145137,
      "grad_norm": 2.375,
      "learning_rate": 0.0004880470553542767,
      "loss": 5.5579,
      "mean_token_accuracy": 0.1599155217409134,
      "num_tokens": 25699255.0,
      "step": 12385
    },
    {
      "entropy": 5.755588674545288,
      "epoch": 1.123911465892598,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048803664734405495,
      "loss": 5.5389,
      "mean_token_accuracy": 0.15058789551258087,
      "num_tokens": 25710714.0,
      "step": 12390
    },
    {
      "entropy": 5.800116968154907,
      "epoch": 1.1243650217706822,
      "grad_norm": 2.515625,
      "learning_rate": 0.000488026234928181,
      "loss": 5.521,
      "mean_token_accuracy": 0.1584481790661812,
      "num_tokens": 25721722.0,
      "step": 12395
    },
    {
      "entropy": 5.7744056224823,
      "epoch": 1.1248185776487662,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004880158181068703,
      "loss": 5.5105,
      "mean_token_accuracy": 0.16380204558372496,
      "num_tokens": 25732096.0,
      "step": 12400
    },
    {
      "entropy": 5.801934719085693,
      "epoch": 1.1252721335268505,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048800539688033807,
      "loss": 5.5733,
      "mean_token_accuracy": 0.16195740401744843,
      "num_tokens": 25741499.0,
      "step": 12405
    },
    {
      "entropy": 5.8076496601104735,
      "epoch": 1.1257256894049348,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004879949712488001,
      "loss": 5.6422,
      "mean_token_accuracy": 0.15778808444738388,
      "num_tokens": 25751913.0,
      "step": 12410
    },
    {
      "entropy": 5.772350072860718,
      "epoch": 1.1261792452830188,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048798454121247177,
      "loss": 5.5817,
      "mean_token_accuracy": 0.15629185736179352,
      "num_tokens": 25762602.0,
      "step": 12415
    },
    {
      "entropy": 5.760409355163574,
      "epoch": 1.126632801161103,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004879741067715688,
      "loss": 5.5413,
      "mean_token_accuracy": 0.16050037890672683,
      "num_tokens": 25773872.0,
      "step": 12420
    },
    {
      "entropy": 5.854652976989746,
      "epoch": 1.1270863570391871,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048796366792630715,
      "loss": 5.5763,
      "mean_token_accuracy": 0.15669984519481658,
      "num_tokens": 25784723.0,
      "step": 12425
    },
    {
      "entropy": 5.795785188674927,
      "epoch": 1.1275399129172714,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004879532246769025,
      "loss": 5.5614,
      "mean_token_accuracy": 0.15872460901737212,
      "num_tokens": 25794897.0,
      "step": 12430
    },
    {
      "entropy": 5.7321044445037845,
      "epoch": 1.1279934687953557,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004879427770235709,
      "loss": 5.4753,
      "mean_token_accuracy": 0.1700018599629402,
      "num_tokens": 25804445.0,
      "step": 12435
    },
    {
      "entropy": 5.843517780303955,
      "epoch": 1.1284470246734397,
      "grad_norm": 3.359375,
      "learning_rate": 0.00048793232496652846,
      "loss": 5.6137,
      "mean_token_accuracy": 0.15143083184957504,
      "num_tokens": 25815495.0,
      "step": 12440
    },
    {
      "entropy": 5.803572082519532,
      "epoch": 1.128900580551524,
      "grad_norm": 2.875,
      "learning_rate": 0.00048792186850599124,
      "loss": 5.5551,
      "mean_token_accuracy": 0.16380234956741332,
      "num_tokens": 25826120.0,
      "step": 12445
    },
    {
      "entropy": 5.700089597702027,
      "epoch": 1.1293541364296082,
      "grad_norm": 3.0,
      "learning_rate": 0.0004879114076421755,
      "loss": 5.5522,
      "mean_token_accuracy": 0.15748713314533233,
      "num_tokens": 25835926.0,
      "step": 12450
    },
    {
      "entropy": 5.763828897476197,
      "epoch": 1.1298076923076923,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048790094237529763,
      "loss": 5.5929,
      "mean_token_accuracy": 0.15242881923913956,
      "num_tokens": 25846095.0,
      "step": 12455
    },
    {
      "entropy": 5.843813228607178,
      "epoch": 1.1302612481857766,
      "grad_norm": 2.828125,
      "learning_rate": 0.000487890472705574,
      "loss": 5.5785,
      "mean_token_accuracy": 0.1516404166817665,
      "num_tokens": 25858541.0,
      "step": 12460
    },
    {
      "entropy": 5.830556011199951,
      "epoch": 1.1307148040638606,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048787999863322113,
      "loss": 5.4803,
      "mean_token_accuracy": 0.16678747832775115,
      "num_tokens": 25867664.0,
      "step": 12465
    },
    {
      "entropy": 5.7513384342193605,
      "epoch": 1.1311683599419449,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004878695201584557,
      "loss": 5.5497,
      "mean_token_accuracy": 0.1595431312918663,
      "num_tokens": 25877300.0,
      "step": 12470
    },
    {
      "entropy": 5.741888809204101,
      "epoch": 1.1316219158200291,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004878590372814943,
      "loss": 5.528,
      "mean_token_accuracy": 0.16952641308307648,
      "num_tokens": 25888009.0,
      "step": 12475
    },
    {
      "entropy": 5.832792472839356,
      "epoch": 1.1320754716981132,
      "grad_norm": 2.5,
      "learning_rate": 0.00048784855000255374,
      "loss": 5.6174,
      "mean_token_accuracy": 0.15054194182157515,
      "num_tokens": 25897285.0,
      "step": 12480
    },
    {
      "entropy": 5.743819046020508,
      "epoch": 1.1325290275761974,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048783805832185097,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16487626284360885,
      "num_tokens": 25907355.0,
      "step": 12485
    },
    {
      "entropy": 5.756371068954468,
      "epoch": 1.1329825834542815,
      "grad_norm": 2.5,
      "learning_rate": 0.0004878275622396029,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16572344452142715,
      "num_tokens": 25918566.0,
      "step": 12490
    },
    {
      "entropy": 5.833869886398316,
      "epoch": 1.1334361393323658,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048781706175602665,
      "loss": 5.6172,
      "mean_token_accuracy": 0.15691329091787337,
      "num_tokens": 25929145.0,
      "step": 12495
    },
    {
      "entropy": 5.709743022918701,
      "epoch": 1.13388969521045,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048780655687133927,
      "loss": 5.4567,
      "mean_token_accuracy": 0.16240618526935577,
      "num_tokens": 25938163.0,
      "step": 12500
    },
    {
      "entropy": 5.722642660140991,
      "epoch": 1.134343251088534,
      "grad_norm": 2.375,
      "learning_rate": 0.00048779604758575806,
      "loss": 5.5321,
      "mean_token_accuracy": 0.15595575273036957,
      "num_tokens": 25948962.0,
      "step": 12505
    },
    {
      "entropy": 5.782497930526733,
      "epoch": 1.1347968069666183,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004877855338995004,
      "loss": 5.4949,
      "mean_token_accuracy": 0.15355829894542694,
      "num_tokens": 25959653.0,
      "step": 12510
    },
    {
      "entropy": 5.850715589523316,
      "epoch": 1.1352503628447024,
      "grad_norm": 2.375,
      "learning_rate": 0.00048777501581278367,
      "loss": 5.5724,
      "mean_token_accuracy": 0.159636926651001,
      "num_tokens": 25970268.0,
      "step": 12515
    },
    {
      "entropy": 5.741877746582031,
      "epoch": 1.1357039187227866,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004877644933258254,
      "loss": 5.5204,
      "mean_token_accuracy": 0.15760474652051926,
      "num_tokens": 25980321.0,
      "step": 12520
    },
    {
      "entropy": 5.701567316055298,
      "epoch": 1.136157474600871,
      "grad_norm": 2.5,
      "learning_rate": 0.0004877539664388432,
      "loss": 5.5312,
      "mean_token_accuracy": 0.15673816055059434,
      "num_tokens": 25992466.0,
      "step": 12525
    },
    {
      "entropy": 5.76544303894043,
      "epoch": 1.136611030478955,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048774343515205464,
      "loss": 5.5296,
      "mean_token_accuracy": 0.16025177836418153,
      "num_tokens": 26003553.0,
      "step": 12530
    },
    {
      "entropy": 5.799270391464233,
      "epoch": 1.1370645863570392,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004877328994656778,
      "loss": 5.5524,
      "mean_token_accuracy": 0.1588145911693573,
      "num_tokens": 26013299.0,
      "step": 12535
    },
    {
      "entropy": 5.783595228195191,
      "epoch": 1.1375181422351233,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004877223593799303,
      "loss": 5.5324,
      "mean_token_accuracy": 0.16661528646945953,
      "num_tokens": 26022913.0,
      "step": 12540
    },
    {
      "entropy": 5.738007736206055,
      "epoch": 1.1379716981132075,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004877118148950302,
      "loss": 5.5268,
      "mean_token_accuracy": 0.16556895524263382,
      "num_tokens": 26033191.0,
      "step": 12545
    },
    {
      "entropy": 5.889295864105224,
      "epoch": 1.1384252539912918,
      "grad_norm": 4.875,
      "learning_rate": 0.00048770126601119555,
      "loss": 5.652,
      "mean_token_accuracy": 0.15249367132782937,
      "num_tokens": 26043517.0,
      "step": 12550
    },
    {
      "entropy": 5.837987804412842,
      "epoch": 1.1388788098693758,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048769071272864454,
      "loss": 5.6224,
      "mean_token_accuracy": 0.1521178126335144,
      "num_tokens": 26054173.0,
      "step": 12555
    },
    {
      "entropy": 5.787218427658081,
      "epoch": 1.13933236574746,
      "grad_norm": 3.375,
      "learning_rate": 0.00048768015504759536,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15482586920261382,
      "num_tokens": 26064929.0,
      "step": 12560
    },
    {
      "entropy": 5.745972681045532,
      "epoch": 1.1397859216255442,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004876695929682665,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16436675935983658,
      "num_tokens": 26074872.0,
      "step": 12565
    },
    {
      "entropy": 5.748572635650635,
      "epoch": 1.1402394775036284,
      "grad_norm": 3.75,
      "learning_rate": 0.00048765902649087605,
      "loss": 5.4376,
      "mean_token_accuracy": 0.16234900653362275,
      "num_tokens": 26086328.0,
      "step": 12570
    },
    {
      "entropy": 5.773342275619507,
      "epoch": 1.1406930333817127,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048764845561564285,
      "loss": 5.5309,
      "mean_token_accuracy": 0.15696113854646682,
      "num_tokens": 26096477.0,
      "step": 12575
    },
    {
      "entropy": 5.744155168533325,
      "epoch": 1.1411465892597967,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048763788034278545,
      "loss": 5.5092,
      "mean_token_accuracy": 0.165939399600029,
      "num_tokens": 26106183.0,
      "step": 12580
    },
    {
      "entropy": 5.742094278335571,
      "epoch": 1.141600145137881,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004876273006725225,
      "loss": 5.5578,
      "mean_token_accuracy": 0.1599455252289772,
      "num_tokens": 26117932.0,
      "step": 12585
    },
    {
      "entropy": 5.734071397781372,
      "epoch": 1.1420537010159653,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048761671660507266,
      "loss": 5.5173,
      "mean_token_accuracy": 0.16329427510499955,
      "num_tokens": 26128616.0,
      "step": 12590
    },
    {
      "entropy": 5.762496709823608,
      "epoch": 1.1425072568940493,
      "grad_norm": 2.265625,
      "learning_rate": 0.000487606128140655,
      "loss": 5.5194,
      "mean_token_accuracy": 0.15964258164167405,
      "num_tokens": 26138758.0,
      "step": 12595
    },
    {
      "entropy": 5.7277239799499515,
      "epoch": 1.1429608127721336,
      "grad_norm": 2.375,
      "learning_rate": 0.0004875955352794885,
      "loss": 5.511,
      "mean_token_accuracy": 0.1583825707435608,
      "num_tokens": 26149199.0,
      "step": 12600
    },
    {
      "entropy": 5.795366382598877,
      "epoch": 1.1434143686502176,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004875849380217921,
      "loss": 5.6151,
      "mean_token_accuracy": 0.1525517076253891,
      "num_tokens": 26160276.0,
      "step": 12605
    },
    {
      "entropy": 5.814324378967285,
      "epoch": 1.1438679245283019,
      "grad_norm": 3.484375,
      "learning_rate": 0.000487574336367785,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1648275911808014,
      "num_tokens": 26169564.0,
      "step": 12610
    },
    {
      "entropy": 5.734959602355957,
      "epoch": 1.1443214804063861,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004875637303176865,
      "loss": 5.4687,
      "mean_token_accuracy": 0.16495396196842194,
      "num_tokens": 26179135.0,
      "step": 12615
    },
    {
      "entropy": 5.836484622955322,
      "epoch": 1.1447750362844702,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048755311987171585,
      "loss": 5.618,
      "mean_token_accuracy": 0.15689624547958375,
      "num_tokens": 26190116.0,
      "step": 12620
    },
    {
      "entropy": 5.788093996047974,
      "epoch": 1.1452285921625545,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048754250503009244,
      "loss": 5.6653,
      "mean_token_accuracy": 0.15636720955371858,
      "num_tokens": 26200847.0,
      "step": 12625
    },
    {
      "entropy": 5.871278953552246,
      "epoch": 1.1456821480406387,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048753188579303596,
      "loss": 5.637,
      "mean_token_accuracy": 0.1531703382730484,
      "num_tokens": 26211151.0,
      "step": 12630
    },
    {
      "entropy": 5.765450811386108,
      "epoch": 1.1461357039187228,
      "grad_norm": 2.875,
      "learning_rate": 0.0004875212621607659,
      "loss": 5.5443,
      "mean_token_accuracy": 0.16738537698984146,
      "num_tokens": 26221273.0,
      "step": 12635
    },
    {
      "entropy": 5.709097051620484,
      "epoch": 1.146589259796807,
      "grad_norm": 3.0,
      "learning_rate": 0.00048751063413350203,
      "loss": 5.6428,
      "mean_token_accuracy": 0.15085833817720412,
      "num_tokens": 26231770.0,
      "step": 12640
    },
    {
      "entropy": 5.804998397827148,
      "epoch": 1.147042815674891,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000487500001711464,
      "loss": 5.6047,
      "mean_token_accuracy": 0.15265530496835708,
      "num_tokens": 26241386.0,
      "step": 12645
    },
    {
      "entropy": 5.930115699768066,
      "epoch": 1.1474963715529753,
      "grad_norm": 3.5,
      "learning_rate": 0.0004874893648948719,
      "loss": 5.6558,
      "mean_token_accuracy": 0.1570476070046425,
      "num_tokens": 26251991.0,
      "step": 12650
    },
    {
      "entropy": 5.884071588516235,
      "epoch": 1.1479499274310596,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048747872368394546,
      "loss": 5.7088,
      "mean_token_accuracy": 0.15459849685430527,
      "num_tokens": 26262331.0,
      "step": 12655
    },
    {
      "entropy": 5.893118381500244,
      "epoch": 1.1484034833091437,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048746807807890496,
      "loss": 5.5946,
      "mean_token_accuracy": 0.15754268020391465,
      "num_tokens": 26272939.0,
      "step": 12660
    },
    {
      "entropy": 5.81418251991272,
      "epoch": 1.148857039187228,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048745742807997046,
      "loss": 5.5351,
      "mean_token_accuracy": 0.1636081039905548,
      "num_tokens": 26282849.0,
      "step": 12665
    },
    {
      "entropy": 5.764125156402588,
      "epoch": 1.149310595065312,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048744677368736216,
      "loss": 5.569,
      "mean_token_accuracy": 0.16500801891088485,
      "num_tokens": 26292449.0,
      "step": 12670
    },
    {
      "entropy": 5.703001880645752,
      "epoch": 1.1497641509433962,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048743611490130044,
      "loss": 5.5198,
      "mean_token_accuracy": 0.15898376256227492,
      "num_tokens": 26302506.0,
      "step": 12675
    },
    {
      "entropy": 5.806844758987427,
      "epoch": 1.1502177068214805,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048742545172200576,
      "loss": 5.5633,
      "mean_token_accuracy": 0.1618367314338684,
      "num_tokens": 26312346.0,
      "step": 12680
    },
    {
      "entropy": 5.864315938949585,
      "epoch": 1.1506712626995645,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004874147841496986,
      "loss": 5.5806,
      "mean_token_accuracy": 0.1587552860379219,
      "num_tokens": 26323173.0,
      "step": 12685
    },
    {
      "entropy": 5.852038478851318,
      "epoch": 1.1511248185776488,
      "grad_norm": 2.125,
      "learning_rate": 0.0004874041121845996,
      "loss": 5.6315,
      "mean_token_accuracy": 0.15016974806785582,
      "num_tokens": 26334371.0,
      "step": 12690
    },
    {
      "entropy": 5.779274749755859,
      "epoch": 1.1515783744557329,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004873934358269294,
      "loss": 5.5714,
      "mean_token_accuracy": 0.1596380278468132,
      "num_tokens": 26344506.0,
      "step": 12695
    },
    {
      "entropy": 5.848462343215942,
      "epoch": 1.1520319303338171,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004873827550769088,
      "loss": 5.5813,
      "mean_token_accuracy": 0.15390973091125487,
      "num_tokens": 26354534.0,
      "step": 12700
    },
    {
      "entropy": 5.8125227928161625,
      "epoch": 1.1524854862119014,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004873720699347587,
      "loss": 5.5591,
      "mean_token_accuracy": 0.15776387751102447,
      "num_tokens": 26365682.0,
      "step": 12705
    },
    {
      "entropy": 5.862371015548706,
      "epoch": 1.1529390420899854,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004873613804007001,
      "loss": 5.6512,
      "mean_token_accuracy": 0.1586606740951538,
      "num_tokens": 26376702.0,
      "step": 12710
    },
    {
      "entropy": 5.741703462600708,
      "epoch": 1.1533925979680697,
      "grad_norm": 2.0,
      "learning_rate": 0.0004873506864749541,
      "loss": 5.4832,
      "mean_token_accuracy": 0.16207378655672072,
      "num_tokens": 26387154.0,
      "step": 12715
    },
    {
      "entropy": 5.889726400375366,
      "epoch": 1.1538461538461537,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004873399881577417,
      "loss": 5.6339,
      "mean_token_accuracy": 0.15240994989871978,
      "num_tokens": 26397883.0,
      "step": 12720
    },
    {
      "entropy": 5.675021457672119,
      "epoch": 1.154299709724238,
      "grad_norm": 6.75,
      "learning_rate": 0.0004873292854492842,
      "loss": 5.4699,
      "mean_token_accuracy": 0.16465118974447251,
      "num_tokens": 26407613.0,
      "step": 12725
    },
    {
      "entropy": 5.6305344581604,
      "epoch": 1.1547532656023223,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004873185783498031,
      "loss": 5.464,
      "mean_token_accuracy": 0.168102265894413,
      "num_tokens": 26417221.0,
      "step": 12730
    },
    {
      "entropy": 5.789429235458374,
      "epoch": 1.1552068214804063,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004873078668595197,
      "loss": 5.5549,
      "mean_token_accuracy": 0.15822538137435913,
      "num_tokens": 26427976.0,
      "step": 12735
    },
    {
      "entropy": 5.840104579925537,
      "epoch": 1.1556603773584906,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048729715097865534,
      "loss": 5.5974,
      "mean_token_accuracy": 0.15557189732789994,
      "num_tokens": 26439213.0,
      "step": 12740
    },
    {
      "entropy": 5.784878301620483,
      "epoch": 1.1561139332365746,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004872864307074319,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1623124361038208,
      "num_tokens": 26449763.0,
      "step": 12745
    },
    {
      "entropy": 5.8193315982818605,
      "epoch": 1.156567489114659,
      "grad_norm": 3.15625,
      "learning_rate": 0.00048727570604607096,
      "loss": 5.6067,
      "mean_token_accuracy": 0.1595538079738617,
      "num_tokens": 26460133.0,
      "step": 12750
    },
    {
      "entropy": 5.805304288864136,
      "epoch": 1.1570210449927432,
      "grad_norm": 2.25,
      "learning_rate": 0.00048726497699479437,
      "loss": 5.5213,
      "mean_token_accuracy": 0.16144013851881028,
      "num_tokens": 26470164.0,
      "step": 12755
    },
    {
      "entropy": 5.751625871658325,
      "epoch": 1.1574746008708272,
      "grad_norm": 3.40625,
      "learning_rate": 0.000487254243553824,
      "loss": 5.5666,
      "mean_token_accuracy": 0.15875484496355058,
      "num_tokens": 26480314.0,
      "step": 12760
    },
    {
      "entropy": 5.69498381614685,
      "epoch": 1.1579281567489115,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004872435057233817,
      "loss": 5.5259,
      "mean_token_accuracy": 0.167743419110775,
      "num_tokens": 26491184.0,
      "step": 12765
    },
    {
      "entropy": 5.720418930053711,
      "epoch": 1.1583817126269957,
      "grad_norm": 2.625,
      "learning_rate": 0.00048723276350368957,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16562064588069916,
      "num_tokens": 26501432.0,
      "step": 12770
    },
    {
      "entropy": 5.793333959579468,
      "epoch": 1.1588352685050798,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048722201689496984,
      "loss": 5.546,
      "mean_token_accuracy": 0.15738675892353057,
      "num_tokens": 26510379.0,
      "step": 12775
    },
    {
      "entropy": 5.772318315505982,
      "epoch": 1.159288824383164,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004872112658974447,
      "loss": 5.5061,
      "mean_token_accuracy": 0.1616644725203514,
      "num_tokens": 26519454.0,
      "step": 12780
    },
    {
      "entropy": 5.709981536865234,
      "epoch": 1.159742380261248,
      "grad_norm": 2.5,
      "learning_rate": 0.00048720051051133657,
      "loss": 5.4611,
      "mean_token_accuracy": 0.16261485368013381,
      "num_tokens": 26531325.0,
      "step": 12785
    },
    {
      "entropy": 5.712454128265381,
      "epoch": 1.1601959361393324,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004871897507368678,
      "loss": 5.568,
      "mean_token_accuracy": 0.15962187498807906,
      "num_tokens": 26541510.0,
      "step": 12790
    },
    {
      "entropy": 5.835890865325927,
      "epoch": 1.1606494920174166,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048717898657426093,
      "loss": 5.7005,
      "mean_token_accuracy": 0.15500900223851205,
      "num_tokens": 26553400.0,
      "step": 12795
    },
    {
      "entropy": 5.850030565261841,
      "epoch": 1.1611030478955007,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048716821802373856,
      "loss": 5.5649,
      "mean_token_accuracy": 0.15857833474874497,
      "num_tokens": 26564340.0,
      "step": 12800
    },
    {
      "entropy": 5.760999774932861,
      "epoch": 1.161556603773585,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048715744508552334,
      "loss": 5.515,
      "mean_token_accuracy": 0.16316769570112227,
      "num_tokens": 26573615.0,
      "step": 12805
    },
    {
      "entropy": 5.788042688369751,
      "epoch": 1.1620101596516692,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004871466677598381,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1539304107427597,
      "num_tokens": 26584149.0,
      "step": 12810
    },
    {
      "entropy": 5.75307559967041,
      "epoch": 1.1624637155297532,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004871358860469057,
      "loss": 5.5773,
      "mean_token_accuracy": 0.15191474556922913,
      "num_tokens": 26595064.0,
      "step": 12815
    },
    {
      "entropy": 5.764725685119629,
      "epoch": 1.1629172714078375,
      "grad_norm": 3.765625,
      "learning_rate": 0.00048712509994694903,
      "loss": 5.5582,
      "mean_token_accuracy": 0.1636511579155922,
      "num_tokens": 26605636.0,
      "step": 12820
    },
    {
      "entropy": 5.807230424880982,
      "epoch": 1.1633708272859216,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004871143094601914,
      "loss": 5.6646,
      "mean_token_accuracy": 0.1558474525809288,
      "num_tokens": 26616065.0,
      "step": 12825
    },
    {
      "entropy": 5.852508878707885,
      "epoch": 1.1638243831640058,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004871035145868557,
      "loss": 5.6006,
      "mean_token_accuracy": 0.15671539455652236,
      "num_tokens": 26627246.0,
      "step": 12830
    },
    {
      "entropy": 5.784663295745849,
      "epoch": 1.16427793904209,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048709271532716524,
      "loss": 5.558,
      "mean_token_accuracy": 0.15833499878644944,
      "num_tokens": 26637630.0,
      "step": 12835
    },
    {
      "entropy": 5.754474306106568,
      "epoch": 1.1647314949201741,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004870819116813434,
      "loss": 5.646,
      "mean_token_accuracy": 0.15717677325010299,
      "num_tokens": 26648683.0,
      "step": 12840
    },
    {
      "entropy": 5.832828712463379,
      "epoch": 1.1651850507982584,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048707110364961363,
      "loss": 5.575,
      "mean_token_accuracy": 0.15657560229301454,
      "num_tokens": 26659573.0,
      "step": 12845
    },
    {
      "entropy": 5.8101729393005375,
      "epoch": 1.1656386066763424,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048706029123219937,
      "loss": 5.484,
      "mean_token_accuracy": 0.16560710817575455,
      "num_tokens": 26669384.0,
      "step": 12850
    },
    {
      "entropy": 5.852849054336548,
      "epoch": 1.1660921625544267,
      "grad_norm": 6.84375,
      "learning_rate": 0.00048704947442932426,
      "loss": 5.6213,
      "mean_token_accuracy": 0.15931399315595626,
      "num_tokens": 26679721.0,
      "step": 12855
    },
    {
      "entropy": 5.837862539291382,
      "epoch": 1.166545718432511,
      "grad_norm": 4.4375,
      "learning_rate": 0.00048703865324121196,
      "loss": 5.5165,
      "mean_token_accuracy": 0.16517172753810883,
      "num_tokens": 26689923.0,
      "step": 12860
    },
    {
      "entropy": 5.735535573959351,
      "epoch": 1.166999274310595,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048702782766808637,
      "loss": 5.5045,
      "mean_token_accuracy": 0.16461092084646226,
      "num_tokens": 26699953.0,
      "step": 12865
    },
    {
      "entropy": 5.819210147857666,
      "epoch": 1.1674528301886793,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004870169977101711,
      "loss": 5.7186,
      "mean_token_accuracy": 0.15123931020498277,
      "num_tokens": 26712099.0,
      "step": 12870
    },
    {
      "entropy": 5.848005580902099,
      "epoch": 1.1679063860667633,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048700616336769037,
      "loss": 5.5708,
      "mean_token_accuracy": 0.1640794813632965,
      "num_tokens": 26722334.0,
      "step": 12875
    },
    {
      "entropy": 5.736310243606567,
      "epoch": 1.1683599419448476,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048699532464086817,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16210989505052567,
      "num_tokens": 26732470.0,
      "step": 12880
    },
    {
      "entropy": 5.74801115989685,
      "epoch": 1.1688134978229319,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048698448152992865,
      "loss": 5.5328,
      "mean_token_accuracy": 0.15769471377134323,
      "num_tokens": 26742015.0,
      "step": 12885
    },
    {
      "entropy": 5.803753805160523,
      "epoch": 1.169267053701016,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048697363403509605,
      "loss": 5.5491,
      "mean_token_accuracy": 0.15741512477397918,
      "num_tokens": 26753231.0,
      "step": 12890
    },
    {
      "entropy": 5.803143548965454,
      "epoch": 1.1697206095791002,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004869627821565946,
      "loss": 5.5326,
      "mean_token_accuracy": 0.15670391768217087,
      "num_tokens": 26763895.0,
      "step": 12895
    },
    {
      "entropy": 5.762263393402099,
      "epoch": 1.1701741654571842,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004869519258946489,
      "loss": 5.6294,
      "mean_token_accuracy": 0.15778597742319106,
      "num_tokens": 26774208.0,
      "step": 12900
    },
    {
      "entropy": 5.802394580841065,
      "epoch": 1.1706277213352685,
      "grad_norm": 3.65625,
      "learning_rate": 0.00048694106524948324,
      "loss": 5.5734,
      "mean_token_accuracy": 0.1610211968421936,
      "num_tokens": 26785554.0,
      "step": 12905
    },
    {
      "entropy": 5.811024761199951,
      "epoch": 1.1710812772133528,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048693020022132243,
      "loss": 5.5116,
      "mean_token_accuracy": 0.16102454885840417,
      "num_tokens": 26795488.0,
      "step": 12910
    },
    {
      "entropy": 5.755540609359741,
      "epoch": 1.1715348330914368,
      "grad_norm": 3.484375,
      "learning_rate": 0.000486919330810391,
      "loss": 5.6397,
      "mean_token_accuracy": 0.15020401924848556,
      "num_tokens": 26806801.0,
      "step": 12915
    },
    {
      "entropy": 5.851099824905395,
      "epoch": 1.171988388969521,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004869084570169139,
      "loss": 5.5605,
      "mean_token_accuracy": 0.16339874565601348,
      "num_tokens": 26816869.0,
      "step": 12920
    },
    {
      "entropy": 5.79552903175354,
      "epoch": 1.172441944847605,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048689757884111585,
      "loss": 5.5401,
      "mean_token_accuracy": 0.16142162531614304,
      "num_tokens": 26827003.0,
      "step": 12925
    },
    {
      "entropy": 5.7379802703857425,
      "epoch": 1.1728955007256894,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048688669628322185,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16071861535310744,
      "num_tokens": 26837065.0,
      "step": 12930
    },
    {
      "entropy": 5.741271305084228,
      "epoch": 1.1733490566037736,
      "grad_norm": 3.0625,
      "learning_rate": 0.000486875809343457,
      "loss": 5.4685,
      "mean_token_accuracy": 0.16105515658855438,
      "num_tokens": 26847487.0,
      "step": 12935
    },
    {
      "entropy": 5.807944583892822,
      "epoch": 1.1738026124818577,
      "grad_norm": 2.0,
      "learning_rate": 0.00048686491802204645,
      "loss": 5.5324,
      "mean_token_accuracy": 0.16912297606468202,
      "num_tokens": 26858407.0,
      "step": 12940
    },
    {
      "entropy": 5.809441232681275,
      "epoch": 1.174256168359942,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004868540223192154,
      "loss": 5.5555,
      "mean_token_accuracy": 0.16272309571504592,
      "num_tokens": 26868469.0,
      "step": 12945
    },
    {
      "entropy": 5.806724882125854,
      "epoch": 1.1747097242380262,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004868431222351892,
      "loss": 5.5972,
      "mean_token_accuracy": 0.15936568677425383,
      "num_tokens": 26879580.0,
      "step": 12950
    },
    {
      "entropy": 5.77673454284668,
      "epoch": 1.1751632801161103,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048683221777019313,
      "loss": 5.5673,
      "mean_token_accuracy": 0.15987883508205414,
      "num_tokens": 26889513.0,
      "step": 12955
    },
    {
      "entropy": 5.8022441387176515,
      "epoch": 1.1756168359941945,
      "grad_norm": 4.3125,
      "learning_rate": 0.00048682130892445293,
      "loss": 5.6529,
      "mean_token_accuracy": 0.16260171681642532,
      "num_tokens": 26900792.0,
      "step": 12960
    },
    {
      "entropy": 5.798989295959473,
      "epoch": 1.1760703918722786,
      "grad_norm": 2.6875,
      "learning_rate": 0.000486810395698194,
      "loss": 5.5344,
      "mean_token_accuracy": 0.16421898007392882,
      "num_tokens": 26911298.0,
      "step": 12965
    },
    {
      "entropy": 5.7976226806640625,
      "epoch": 1.1765239477503628,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048679947809164223,
      "loss": 5.604,
      "mean_token_accuracy": 0.15284128934144975,
      "num_tokens": 26921545.0,
      "step": 12970
    },
    {
      "entropy": 5.820248126983643,
      "epoch": 1.176977503628447,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004867885561050232,
      "loss": 5.5466,
      "mean_token_accuracy": 0.16871648132801056,
      "num_tokens": 26932140.0,
      "step": 12975
    },
    {
      "entropy": 5.809496974945068,
      "epoch": 1.1774310595065312,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004867776297385629,
      "loss": 5.5833,
      "mean_token_accuracy": 0.1564341276884079,
      "num_tokens": 26942634.0,
      "step": 12980
    },
    {
      "entropy": 5.786431074142456,
      "epoch": 1.1778846153846154,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048676669899248723,
      "loss": 5.5656,
      "mean_token_accuracy": 0.1618596062064171,
      "num_tokens": 26952758.0,
      "step": 12985
    },
    {
      "entropy": 5.722311353683471,
      "epoch": 1.1783381712626997,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004867557638670222,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1648661747574806,
      "num_tokens": 26963439.0,
      "step": 12990
    },
    {
      "entropy": 5.763740205764771,
      "epoch": 1.1787917271407837,
      "grad_norm": 3.125,
      "learning_rate": 0.0004867448243623941,
      "loss": 5.5437,
      "mean_token_accuracy": 0.15486639440059663,
      "num_tokens": 26974331.0,
      "step": 12995
    },
    {
      "entropy": 5.7536359310150145,
      "epoch": 1.179245283018868,
      "grad_norm": 3.625,
      "learning_rate": 0.000486733880478829,
      "loss": 5.4275,
      "mean_token_accuracy": 0.1671615555882454,
      "num_tokens": 26983416.0,
      "step": 13000
    },
    {
      "entropy": 5.73840594291687,
      "epoch": 1.179698838896952,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048672293221655333,
      "loss": 5.5485,
      "mean_token_accuracy": 0.1658090427517891,
      "num_tokens": 26994176.0,
      "step": 13005
    },
    {
      "entropy": 5.763584995269776,
      "epoch": 1.1801523947750363,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004867119795757935,
      "loss": 5.5498,
      "mean_token_accuracy": 0.15904117971658707,
      "num_tokens": 27003295.0,
      "step": 13010
    },
    {
      "entropy": 5.775843000411987,
      "epoch": 1.1806059506531206,
      "grad_norm": 3.671875,
      "learning_rate": 0.000486701022556776,
      "loss": 5.5169,
      "mean_token_accuracy": 0.15969693511724473,
      "num_tokens": 27014018.0,
      "step": 13015
    },
    {
      "entropy": 5.775488948822021,
      "epoch": 1.1810595065312046,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004866900611597273,
      "loss": 5.537,
      "mean_token_accuracy": 0.16250109523534775,
      "num_tokens": 27024706.0,
      "step": 13020
    },
    {
      "entropy": 5.770761013031006,
      "epoch": 1.1815130624092889,
      "grad_norm": 8.6875,
      "learning_rate": 0.00048667909538487427,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16427334100008012,
      "num_tokens": 27035801.0,
      "step": 13025
    },
    {
      "entropy": 5.651213026046753,
      "epoch": 1.181966618287373,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004866681252324436,
      "loss": 5.4416,
      "mean_token_accuracy": 0.16571701020002366,
      "num_tokens": 27046596.0,
      "step": 13030
    },
    {
      "entropy": 5.7422198295593265,
      "epoch": 1.1824201741654572,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048665715070266214,
      "loss": 5.5153,
      "mean_token_accuracy": 0.16186168193817138,
      "num_tokens": 27057057.0,
      "step": 13035
    },
    {
      "entropy": 5.707665729522705,
      "epoch": 1.1828737300435415,
      "grad_norm": 3.75,
      "learning_rate": 0.00048664617179575685,
      "loss": 5.5091,
      "mean_token_accuracy": 0.16573376804590226,
      "num_tokens": 27067947.0,
      "step": 13040
    },
    {
      "entropy": 5.782701301574707,
      "epoch": 1.1833272859216255,
      "grad_norm": 3.125,
      "learning_rate": 0.0004866351885119548,
      "loss": 5.5426,
      "mean_token_accuracy": 0.15837194323539733,
      "num_tokens": 27077503.0,
      "step": 13045
    },
    {
      "entropy": 5.851653718948365,
      "epoch": 1.1837808417997098,
      "grad_norm": 2.625,
      "learning_rate": 0.0004866242008514832,
      "loss": 5.672,
      "mean_token_accuracy": 0.15832241624593735,
      "num_tokens": 27088298.0,
      "step": 13050
    },
    {
      "entropy": 5.77230863571167,
      "epoch": 1.1842343976777938,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048661320881456904,
      "loss": 5.5019,
      "mean_token_accuracy": 0.15720409899950027,
      "num_tokens": 27097841.0,
      "step": 13055
    },
    {
      "entropy": 5.698291301727295,
      "epoch": 1.184687953555878,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004866022124014399,
      "loss": 5.4543,
      "mean_token_accuracy": 0.16975428462028502,
      "num_tokens": 27108263.0,
      "step": 13060
    },
    {
      "entropy": 5.7329010486602785,
      "epoch": 1.1851415094339623,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048659121161232313,
      "loss": 5.5408,
      "mean_token_accuracy": 0.16036011278629303,
      "num_tokens": 27119339.0,
      "step": 13065
    },
    {
      "entropy": 5.762398433685303,
      "epoch": 1.1855950653120464,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004865802064474461,
      "loss": 5.5775,
      "mean_token_accuracy": 0.15528168231248857,
      "num_tokens": 27130393.0,
      "step": 13070
    },
    {
      "entropy": 5.853636074066162,
      "epoch": 1.1860486211901307,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048656919690703653,
      "loss": 5.5621,
      "mean_token_accuracy": 0.15676342099905013,
      "num_tokens": 27141156.0,
      "step": 13075
    },
    {
      "entropy": 5.853552293777466,
      "epoch": 1.1865021770682147,
      "grad_norm": 2.4375,
      "learning_rate": 0.000486558182991322,
      "loss": 5.6421,
      "mean_token_accuracy": 0.15721183270215988,
      "num_tokens": 27151630.0,
      "step": 13080
    },
    {
      "entropy": 5.803340864181519,
      "epoch": 1.186955732946299,
      "grad_norm": 3.5,
      "learning_rate": 0.0004865471647005304,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16862793266773224,
      "num_tokens": 27160734.0,
      "step": 13085
    },
    {
      "entropy": 5.772292280197144,
      "epoch": 1.1874092888243832,
      "grad_norm": 4.0,
      "learning_rate": 0.00048653614203488947,
      "loss": 5.5527,
      "mean_token_accuracy": 0.15969373285770416,
      "num_tokens": 27171584.0,
      "step": 13090
    },
    {
      "entropy": 5.705093908309936,
      "epoch": 1.1878628447024673,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004865251149946273,
      "loss": 5.4833,
      "mean_token_accuracy": 0.16997457891702653,
      "num_tokens": 27180817.0,
      "step": 13095
    },
    {
      "entropy": 5.76951847076416,
      "epoch": 1.1883164005805515,
      "grad_norm": 3.8125,
      "learning_rate": 0.00048651408357997173,
      "loss": 5.5472,
      "mean_token_accuracy": 0.15950145721435546,
      "num_tokens": 27190363.0,
      "step": 13100
    },
    {
      "entropy": 5.797454881668091,
      "epoch": 1.1887699564586356,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048650304779115114,
      "loss": 5.5346,
      "mean_token_accuracy": 0.16130672842264177,
      "num_tokens": 27202181.0,
      "step": 13105
    },
    {
      "entropy": 5.74872088432312,
      "epoch": 1.1892235123367199,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004864920076283935,
      "loss": 5.5557,
      "mean_token_accuracy": 0.15868694186210633,
      "num_tokens": 27212514.0,
      "step": 13110
    },
    {
      "entropy": 5.791923332214355,
      "epoch": 1.1896770682148041,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004864809630919273,
      "loss": 5.5457,
      "mean_token_accuracy": 0.16472710520029069,
      "num_tokens": 27223352.0,
      "step": 13115
    },
    {
      "entropy": 5.784434175491333,
      "epoch": 1.1901306240928882,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004864699141819809,
      "loss": 5.4844,
      "mean_token_accuracy": 0.16784779578447342,
      "num_tokens": 27232643.0,
      "step": 13120
    },
    {
      "entropy": 5.703444099426269,
      "epoch": 1.1905841799709724,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048645886089878276,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16358083933591844,
      "num_tokens": 27243793.0,
      "step": 13125
    },
    {
      "entropy": 5.744236326217651,
      "epoch": 1.1910377358490567,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048644780324256153,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16556206792593003,
      "num_tokens": 27254554.0,
      "step": 13130
    },
    {
      "entropy": 5.817253971099854,
      "epoch": 1.1914912917271407,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004864367412135458,
      "loss": 5.5942,
      "mean_token_accuracy": 0.1576319694519043,
      "num_tokens": 27265842.0,
      "step": 13135
    },
    {
      "entropy": 5.8717255115509035,
      "epoch": 1.191944847605225,
      "grad_norm": 4.03125,
      "learning_rate": 0.00048642567481196446,
      "loss": 5.6002,
      "mean_token_accuracy": 0.15470883399248123,
      "num_tokens": 27277617.0,
      "step": 13140
    },
    {
      "entropy": 5.782086992263794,
      "epoch": 1.192398403483309,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004864146040380462,
      "loss": 5.5215,
      "mean_token_accuracy": 0.1639416515827179,
      "num_tokens": 27287450.0,
      "step": 13145
    },
    {
      "entropy": 5.769864892959594,
      "epoch": 1.1928519593613933,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004864035288920202,
      "loss": 5.5358,
      "mean_token_accuracy": 0.1597059339284897,
      "num_tokens": 27297010.0,
      "step": 13150
    },
    {
      "entropy": 5.727917671203613,
      "epoch": 1.1933055152394776,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048639244937411516,
      "loss": 5.5693,
      "mean_token_accuracy": 0.15701576918363572,
      "num_tokens": 27306874.0,
      "step": 13155
    },
    {
      "entropy": 5.833353662490845,
      "epoch": 1.1937590711175616,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048638136548456045,
      "loss": 5.6258,
      "mean_token_accuracy": 0.15229152590036393,
      "num_tokens": 27318762.0,
      "step": 13160
    },
    {
      "entropy": 5.7694495677947994,
      "epoch": 1.194212626995646,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048637027722358536,
      "loss": 5.4385,
      "mean_token_accuracy": 0.17290839701890945,
      "num_tokens": 27327231.0,
      "step": 13165
    },
    {
      "entropy": 5.716942930221558,
      "epoch": 1.1946661828737302,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048635918459141893,
      "loss": 5.4835,
      "mean_token_accuracy": 0.16239961087703705,
      "num_tokens": 27338517.0,
      "step": 13170
    },
    {
      "entropy": 5.7653703689575195,
      "epoch": 1.1951197387518142,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048634808758829073,
      "loss": 5.5505,
      "mean_token_accuracy": 0.16109321564435958,
      "num_tokens": 27348861.0,
      "step": 13175
    },
    {
      "entropy": 5.778063011169434,
      "epoch": 1.1955732946298985,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048633698621443024,
      "loss": 5.5021,
      "mean_token_accuracy": 0.16753173619508743,
      "num_tokens": 27358610.0,
      "step": 13180
    },
    {
      "entropy": 5.762907838821411,
      "epoch": 1.1960268505079825,
      "grad_norm": 2.78125,
      "learning_rate": 0.000486325880470067,
      "loss": 5.6024,
      "mean_token_accuracy": 0.16014077961444856,
      "num_tokens": 27369707.0,
      "step": 13185
    },
    {
      "entropy": 5.768137502670288,
      "epoch": 1.1964804063860668,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004863147703554307,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16034825295209884,
      "num_tokens": 27381030.0,
      "step": 13190
    },
    {
      "entropy": 5.80393877029419,
      "epoch": 1.196933962264151,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048630365587075115,
      "loss": 5.5728,
      "mean_token_accuracy": 0.15652906149625778,
      "num_tokens": 27390866.0,
      "step": 13195
    },
    {
      "entropy": 5.779042196273804,
      "epoch": 1.197387518142235,
      "grad_norm": 3.75,
      "learning_rate": 0.00048629253701625816,
      "loss": 5.4907,
      "mean_token_accuracy": 0.1675584301352501,
      "num_tokens": 27400724.0,
      "step": 13200
    },
    {
      "entropy": 5.760359001159668,
      "epoch": 1.1978410740203194,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004862814137921816,
      "loss": 5.5277,
      "mean_token_accuracy": 0.16160324811935425,
      "num_tokens": 27411212.0,
      "step": 13205
    },
    {
      "entropy": 5.769501209259033,
      "epoch": 1.1982946298984034,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048627028619875157,
      "loss": 5.5583,
      "mean_token_accuracy": 0.1600644275546074,
      "num_tokens": 27421279.0,
      "step": 13210
    },
    {
      "entropy": 5.7910521030426025,
      "epoch": 1.1987481857764877,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004862591542361982,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1610708087682724,
      "num_tokens": 27432520.0,
      "step": 13215
    },
    {
      "entropy": 5.731895923614502,
      "epoch": 1.199201741654572,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004862480179047516,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16615597456693648,
      "num_tokens": 27442246.0,
      "step": 13220
    },
    {
      "entropy": 5.759058856964112,
      "epoch": 1.199655297532656,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004862368772046423,
      "loss": 5.4996,
      "mean_token_accuracy": 0.16290442943572997,
      "num_tokens": 27452134.0,
      "step": 13225
    },
    {
      "entropy": 5.75121111869812,
      "epoch": 1.2001088534107403,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048622573213610057,
      "loss": 5.5526,
      "mean_token_accuracy": 0.15651463568210602,
      "num_tokens": 27462253.0,
      "step": 13230
    },
    {
      "entropy": 5.777753162384033,
      "epoch": 1.2005624092888243,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048621458269935676,
      "loss": 5.5046,
      "mean_token_accuracy": 0.16076795905828475,
      "num_tokens": 27473131.0,
      "step": 13235
    },
    {
      "entropy": 5.767009639739991,
      "epoch": 1.2010159651669086,
      "grad_norm": 4.15625,
      "learning_rate": 0.00048620342889464153,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16573919504880905,
      "num_tokens": 27483683.0,
      "step": 13240
    },
    {
      "entropy": 5.7502676486969,
      "epoch": 1.2014695210449928,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048619227072218565,
      "loss": 5.5077,
      "mean_token_accuracy": 0.15661614686250686,
      "num_tokens": 27493679.0,
      "step": 13245
    },
    {
      "entropy": 5.746943759918213,
      "epoch": 1.2019230769230769,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004861811081822198,
      "loss": 5.5199,
      "mean_token_accuracy": 0.16417332887649536,
      "num_tokens": 27503897.0,
      "step": 13250
    },
    {
      "entropy": 5.796412324905395,
      "epoch": 1.2023766328011611,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004861699412749748,
      "loss": 5.5843,
      "mean_token_accuracy": 0.16416144967079163,
      "num_tokens": 27514560.0,
      "step": 13255
    },
    {
      "entropy": 5.7397260665893555,
      "epoch": 1.2028301886792452,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004861587700006816,
      "loss": 5.4909,
      "mean_token_accuracy": 0.16245728284120559,
      "num_tokens": 27525594.0,
      "step": 13260
    },
    {
      "entropy": 5.706835174560547,
      "epoch": 1.2032837445573294,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048614759435957114,
      "loss": 5.4609,
      "mean_token_accuracy": 0.16556266695261002,
      "num_tokens": 27535676.0,
      "step": 13265
    },
    {
      "entropy": 5.699453735351563,
      "epoch": 1.2037373004354137,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004861364143518747,
      "loss": 5.4515,
      "mean_token_accuracy": 0.15577304512262344,
      "num_tokens": 27545520.0,
      "step": 13270
    },
    {
      "entropy": 5.860692167282105,
      "epoch": 1.2041908563134978,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004861252299778234,
      "loss": 5.5751,
      "mean_token_accuracy": 0.15455857515335084,
      "num_tokens": 27556866.0,
      "step": 13275
    },
    {
      "entropy": 5.767722034454346,
      "epoch": 1.204644412191582,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048611404123764855,
      "loss": 5.5058,
      "mean_token_accuracy": 0.16126000881195068,
      "num_tokens": 27566951.0,
      "step": 13280
    },
    {
      "entropy": 5.768840885162353,
      "epoch": 1.205097968069666,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048610284813158154,
      "loss": 5.5974,
      "mean_token_accuracy": 0.16187702715396882,
      "num_tokens": 27576796.0,
      "step": 13285
    },
    {
      "entropy": 5.691097688674927,
      "epoch": 1.2055515239477503,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004860916506598538,
      "loss": 5.4668,
      "mean_token_accuracy": 0.16215330809354783,
      "num_tokens": 27586902.0,
      "step": 13290
    },
    {
      "entropy": 5.852893590927124,
      "epoch": 1.2060050798258346,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048608044882269695,
      "loss": 5.6794,
      "mean_token_accuracy": 0.1502825379371643,
      "num_tokens": 27599232.0,
      "step": 13295
    },
    {
      "entropy": 5.7961523056030275,
      "epoch": 1.2064586357039186,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004860692426203427,
      "loss": 5.4619,
      "mean_token_accuracy": 0.16296741217374802,
      "num_tokens": 27609022.0,
      "step": 13300
    },
    {
      "entropy": 5.802973508834839,
      "epoch": 1.206912191582003,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004860580320530226,
      "loss": 5.6072,
      "mean_token_accuracy": 0.16079449653625488,
      "num_tokens": 27619829.0,
      "step": 13305
    },
    {
      "entropy": 5.790474843978882,
      "epoch": 1.2073657474600872,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004860468171209687,
      "loss": 5.6109,
      "mean_token_accuracy": 0.1681277260184288,
      "num_tokens": 27629337.0,
      "step": 13310
    },
    {
      "entropy": 5.8300903797149655,
      "epoch": 1.2078193033381712,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004860355978244128,
      "loss": 5.6245,
      "mean_token_accuracy": 0.15967096984386445,
      "num_tokens": 27639377.0,
      "step": 13315
    },
    {
      "entropy": 5.764687156677246,
      "epoch": 1.2082728592162555,
      "grad_norm": 2.71875,
      "learning_rate": 0.000486024374163587,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16351540982723237,
      "num_tokens": 27649989.0,
      "step": 13320
    },
    {
      "entropy": 5.7667320728302,
      "epoch": 1.2087264150943395,
      "grad_norm": 2.75,
      "learning_rate": 0.00048601314613872334,
      "loss": 5.5781,
      "mean_token_accuracy": 0.15575842559337616,
      "num_tokens": 27659967.0,
      "step": 13325
    },
    {
      "entropy": 5.800278425216675,
      "epoch": 1.2091799709724238,
      "grad_norm": 3.34375,
      "learning_rate": 0.000486001913750054,
      "loss": 5.5365,
      "mean_token_accuracy": 0.16597909480333328,
      "num_tokens": 27670099.0,
      "step": 13330
    },
    {
      "entropy": 5.766505098342895,
      "epoch": 1.209633526850508,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004859906769978113,
      "loss": 5.5776,
      "mean_token_accuracy": 0.1558314934372902,
      "num_tokens": 27681497.0,
      "step": 13335
    },
    {
      "entropy": 5.808494424819946,
      "epoch": 1.2100870827285921,
      "grad_norm": 3.5,
      "learning_rate": 0.00048597943588222776,
      "loss": 5.6774,
      "mean_token_accuracy": 0.1571805089712143,
      "num_tokens": 27691828.0,
      "step": 13340
    },
    {
      "entropy": 5.80404691696167,
      "epoch": 1.2105406386066764,
      "grad_norm": 2.125,
      "learning_rate": 0.00048596819040353564,
      "loss": 5.4963,
      "mean_token_accuracy": 0.16271981224417686,
      "num_tokens": 27701426.0,
      "step": 13345
    },
    {
      "entropy": 5.847316646575928,
      "epoch": 1.2109941944847606,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004859569405619676,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1663869321346283,
      "num_tokens": 27711112.0,
      "step": 13350
    },
    {
      "entropy": 5.73911566734314,
      "epoch": 1.2114477503628447,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004859456863577562,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1577397868037224,
      "num_tokens": 27723160.0,
      "step": 13355
    },
    {
      "entropy": 5.819460439682007,
      "epoch": 1.211901306240929,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048593442779113437,
      "loss": 5.5586,
      "mean_token_accuracy": 0.15461450144648553,
      "num_tokens": 27733143.0,
      "step": 13360
    },
    {
      "entropy": 5.793652629852295,
      "epoch": 1.212354862119013,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004859231648623348,
      "loss": 5.5867,
      "mean_token_accuracy": 0.16012637317180634,
      "num_tokens": 27743956.0,
      "step": 13365
    },
    {
      "entropy": 5.774345397949219,
      "epoch": 1.2128084179970973,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048591189757159035,
      "loss": 5.5252,
      "mean_token_accuracy": 0.16550300866365433,
      "num_tokens": 27754192.0,
      "step": 13370
    },
    {
      "entropy": 5.758378171920777,
      "epoch": 1.2132619738751815,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004859006259191342,
      "loss": 5.5849,
      "mean_token_accuracy": 0.15404899418354034,
      "num_tokens": 27763939.0,
      "step": 13375
    },
    {
      "entropy": 5.789527750015258,
      "epoch": 1.2137155297532656,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004858893499051993,
      "loss": 5.5488,
      "mean_token_accuracy": 0.16654455661773682,
      "num_tokens": 27774743.0,
      "step": 13380
    },
    {
      "entropy": 5.790659618377686,
      "epoch": 1.2141690856313498,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004858780695300189,
      "loss": 5.5241,
      "mean_token_accuracy": 0.16331542134284974,
      "num_tokens": 27785344.0,
      "step": 13385
    },
    {
      "entropy": 5.768062257766724,
      "epoch": 1.2146226415094339,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004858667847938263,
      "loss": 5.6107,
      "mean_token_accuracy": 0.15832710415124893,
      "num_tokens": 27797101.0,
      "step": 13390
    },
    {
      "entropy": 5.6932501792907715,
      "epoch": 1.2150761973875182,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048585549569685486,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1687215656042099,
      "num_tokens": 27807581.0,
      "step": 13395
    },
    {
      "entropy": 5.73537974357605,
      "epoch": 1.2155297532656024,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048584420223933805,
      "loss": 5.4595,
      "mean_token_accuracy": 0.16422586292028427,
      "num_tokens": 27818170.0,
      "step": 13400
    },
    {
      "entropy": 5.726855134963989,
      "epoch": 1.2159833091436865,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004858329044215094,
      "loss": 5.5051,
      "mean_token_accuracy": 0.16495947241783143,
      "num_tokens": 27828153.0,
      "step": 13405
    },
    {
      "entropy": 5.8071067333221436,
      "epoch": 1.2164368650217707,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048582160224360253,
      "loss": 5.441,
      "mean_token_accuracy": 0.16480378657579423,
      "num_tokens": 27837681.0,
      "step": 13410
    },
    {
      "entropy": 5.765391826629639,
      "epoch": 1.2168904208998548,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004858102957058513,
      "loss": 5.5589,
      "mean_token_accuracy": 0.16153258532285691,
      "num_tokens": 27847447.0,
      "step": 13415
    },
    {
      "entropy": 5.717009258270264,
      "epoch": 1.217343976777939,
      "grad_norm": 2.125,
      "learning_rate": 0.00048579898480848933,
      "loss": 5.4412,
      "mean_token_accuracy": 0.1613820970058441,
      "num_tokens": 27858586.0,
      "step": 13420
    },
    {
      "entropy": 5.726960611343384,
      "epoch": 1.2177975326560233,
      "grad_norm": 2.5,
      "learning_rate": 0.0004857876695517507,
      "loss": 5.5196,
      "mean_token_accuracy": 0.1647893249988556,
      "num_tokens": 27868832.0,
      "step": 13425
    },
    {
      "entropy": 5.767963409423828,
      "epoch": 1.2182510885341074,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048577634993586923,
      "loss": 5.5564,
      "mean_token_accuracy": 0.15944417864084243,
      "num_tokens": 27879339.0,
      "step": 13430
    },
    {
      "entropy": 5.788715171813965,
      "epoch": 1.2187046444121916,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048576502596107924,
      "loss": 5.5562,
      "mean_token_accuracy": 0.16190617829561232,
      "num_tokens": 27889964.0,
      "step": 13435
    },
    {
      "entropy": 5.831266260147094,
      "epoch": 1.2191582002902757,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004857536976276148,
      "loss": 5.601,
      "mean_token_accuracy": 0.1610369086265564,
      "num_tokens": 27901004.0,
      "step": 13440
    },
    {
      "entropy": 5.770554780960083,
      "epoch": 1.21961175616836,
      "grad_norm": 3.375,
      "learning_rate": 0.0004857423649357102,
      "loss": 5.5119,
      "mean_token_accuracy": 0.16179349422454833,
      "num_tokens": 27910788.0,
      "step": 13445
    },
    {
      "entropy": 5.7940497398376465,
      "epoch": 1.2200653120464442,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048573102788559974,
      "loss": 5.5541,
      "mean_token_accuracy": 0.16251745969057083,
      "num_tokens": 27921360.0,
      "step": 13450
    },
    {
      "entropy": 5.719580793380738,
      "epoch": 1.2205188679245282,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004857196864775179,
      "loss": 5.5048,
      "mean_token_accuracy": 0.16737589687108995,
      "num_tokens": 27931889.0,
      "step": 13455
    },
    {
      "entropy": 5.736418104171753,
      "epoch": 1.2209724238026125,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048570834071169926,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1658502921462059,
      "num_tokens": 27942049.0,
      "step": 13460
    },
    {
      "entropy": 5.829206228256226,
      "epoch": 1.2214259796806965,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048569699058837856,
      "loss": 5.6259,
      "mean_token_accuracy": 0.15900640189647675,
      "num_tokens": 27952416.0,
      "step": 13465
    },
    {
      "entropy": 5.729955196380615,
      "epoch": 1.2218795355587808,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004856856361077903,
      "loss": 5.4995,
      "mean_token_accuracy": 0.1637960731983185,
      "num_tokens": 27962009.0,
      "step": 13470
    },
    {
      "entropy": 5.776455593109131,
      "epoch": 1.222333091436865,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004856742772701694,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1619058683514595,
      "num_tokens": 27972175.0,
      "step": 13475
    },
    {
      "entropy": 5.733641767501831,
      "epoch": 1.2227866473149491,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048566291407575073,
      "loss": 5.5699,
      "mean_token_accuracy": 0.16602234095335006,
      "num_tokens": 27982899.0,
      "step": 13480
    },
    {
      "entropy": 5.7740381240844725,
      "epoch": 1.2232402031930334,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048565154652476937,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16384897232055665,
      "num_tokens": 27993231.0,
      "step": 13485
    },
    {
      "entropy": 5.742705965042115,
      "epoch": 1.2236937590711174,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004856401746174603,
      "loss": 5.4623,
      "mean_token_accuracy": 0.16265252679586412,
      "num_tokens": 28003623.0,
      "step": 13490
    },
    {
      "entropy": 5.7740149974823,
      "epoch": 1.2241473149492017,
      "grad_norm": 4.625,
      "learning_rate": 0.00048562879835405874,
      "loss": 5.4614,
      "mean_token_accuracy": 0.16348905712366105,
      "num_tokens": 28013878.0,
      "step": 13495
    },
    {
      "entropy": 5.709716081619263,
      "epoch": 1.224600870827286,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004856174177348,
      "loss": 5.547,
      "mean_token_accuracy": 0.16888151317834854,
      "num_tokens": 28024224.0,
      "step": 13500
    },
    {
      "entropy": 5.7263744354248045,
      "epoch": 1.22505442670537,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004856060327599194,
      "loss": 5.4671,
      "mean_token_accuracy": 0.1604592740535736,
      "num_tokens": 28034819.0,
      "step": 13505
    },
    {
      "entropy": 5.752088785171509,
      "epoch": 1.2255079825834543,
      "grad_norm": 2.625,
      "learning_rate": 0.00048559464342965233,
      "loss": 5.5453,
      "mean_token_accuracy": 0.1564328759908676,
      "num_tokens": 28044240.0,
      "step": 13510
    },
    {
      "entropy": 5.86681056022644,
      "epoch": 1.2259615384615385,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004855832497442344,
      "loss": 5.6697,
      "mean_token_accuracy": 0.15296784043312073,
      "num_tokens": 28055046.0,
      "step": 13515
    },
    {
      "entropy": 5.731367349624634,
      "epoch": 1.2264150943396226,
      "grad_norm": 2.375,
      "learning_rate": 0.0004855718517039012,
      "loss": 5.4821,
      "mean_token_accuracy": 0.1621873989701271,
      "num_tokens": 28065272.0,
      "step": 13520
    },
    {
      "entropy": 5.813319492340088,
      "epoch": 1.2268686502177069,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048556044930888844,
      "loss": 5.477,
      "mean_token_accuracy": 0.16950096935033798,
      "num_tokens": 28076255.0,
      "step": 13525
    },
    {
      "entropy": 5.782713985443115,
      "epoch": 1.2273222060957911,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004855490425594319,
      "loss": 5.572,
      "mean_token_accuracy": 0.1638113185763359,
      "num_tokens": 28086258.0,
      "step": 13530
    },
    {
      "entropy": 5.79962944984436,
      "epoch": 1.2277757619738752,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004855376314557675,
      "loss": 5.5015,
      "mean_token_accuracy": 0.16507641822099686,
      "num_tokens": 28096497.0,
      "step": 13535
    },
    {
      "entropy": 5.750047874450684,
      "epoch": 1.2282293178519594,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004855262159981313,
      "loss": 5.5325,
      "mean_token_accuracy": 0.1651751533150673,
      "num_tokens": 28106282.0,
      "step": 13540
    },
    {
      "entropy": 5.654364347457886,
      "epoch": 1.2286828737300435,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048551479618675925,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1635674089193344,
      "num_tokens": 28115771.0,
      "step": 13545
    },
    {
      "entropy": 5.859809017181396,
      "epoch": 1.2291364296081277,
      "grad_norm": 3.546875,
      "learning_rate": 0.00048550337202188756,
      "loss": 5.5091,
      "mean_token_accuracy": 0.16698596030473709,
      "num_tokens": 28125477.0,
      "step": 13550
    },
    {
      "entropy": 5.932044172286988,
      "epoch": 1.229589985486212,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004854919435037525,
      "loss": 5.5287,
      "mean_token_accuracy": 0.1602046236395836,
      "num_tokens": 28135800.0,
      "step": 13555
    },
    {
      "entropy": 5.752037429809571,
      "epoch": 1.230043541364296,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004854805106325904,
      "loss": 5.4224,
      "mean_token_accuracy": 0.16074625998735428,
      "num_tokens": 28146779.0,
      "step": 13560
    },
    {
      "entropy": 5.749845886230469,
      "epoch": 1.2304970972423803,
      "grad_norm": 3.0,
      "learning_rate": 0.0004854690734086378,
      "loss": 5.5743,
      "mean_token_accuracy": 0.15834655612707138,
      "num_tokens": 28156819.0,
      "step": 13565
    },
    {
      "entropy": 5.794097423553467,
      "epoch": 1.2309506531204644,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004854576318321311,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1604808062314987,
      "num_tokens": 28168553.0,
      "step": 13570
    },
    {
      "entropy": 5.802403974533081,
      "epoch": 1.2314042089985486,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048544618590330685,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16358250081539155,
      "num_tokens": 28178064.0,
      "step": 13575
    },
    {
      "entropy": 5.814314985275269,
      "epoch": 1.231857764876633,
      "grad_norm": 3.15625,
      "learning_rate": 0.00048543473562240194,
      "loss": 5.6534,
      "mean_token_accuracy": 0.1619192525744438,
      "num_tokens": 28188580.0,
      "step": 13580
    },
    {
      "entropy": 5.767128801345825,
      "epoch": 1.232311320754717,
      "grad_norm": 2.421875,
      "learning_rate": 0.000485423280989653,
      "loss": 5.6031,
      "mean_token_accuracy": 0.15844668000936507,
      "num_tokens": 28199955.0,
      "step": 13585
    },
    {
      "entropy": 5.802313566207886,
      "epoch": 1.2327648766328012,
      "grad_norm": 3.84375,
      "learning_rate": 0.000485411822005297,
      "loss": 5.5805,
      "mean_token_accuracy": 0.15959620401263236,
      "num_tokens": 28210506.0,
      "step": 13590
    },
    {
      "entropy": 5.789461660385132,
      "epoch": 1.2332184325108853,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048540035866957097,
      "loss": 5.5239,
      "mean_token_accuracy": 0.1683552786707878,
      "num_tokens": 28220486.0,
      "step": 13595
    },
    {
      "entropy": 5.830734491348267,
      "epoch": 1.2336719883889695,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004853888909827118,
      "loss": 5.631,
      "mean_token_accuracy": 0.15113798528909683,
      "num_tokens": 28232668.0,
      "step": 13600
    },
    {
      "entropy": 5.777379179000855,
      "epoch": 1.2341255442670538,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004853774189449568,
      "loss": 5.5778,
      "mean_token_accuracy": 0.15810663849115372,
      "num_tokens": 28243733.0,
      "step": 13605
    },
    {
      "entropy": 5.877877855300904,
      "epoch": 1.2345791001451378,
      "grad_norm": 4.125,
      "learning_rate": 0.00048536594255654315,
      "loss": 5.6059,
      "mean_token_accuracy": 0.16085377782583238,
      "num_tokens": 28254953.0,
      "step": 13610
    },
    {
      "entropy": 5.813952350616455,
      "epoch": 1.235032656023222,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004853544618177083,
      "loss": 5.5383,
      "mean_token_accuracy": 0.16349480748176576,
      "num_tokens": 28264959.0,
      "step": 13615
    },
    {
      "entropy": 5.775934934616089,
      "epoch": 1.2354862119013061,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048534297672868945,
      "loss": 5.5321,
      "mean_token_accuracy": 0.16197028309106826,
      "num_tokens": 28275509.0,
      "step": 13620
    },
    {
      "entropy": 5.7924230098724365,
      "epoch": 1.2359397677793904,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048533148728972434,
      "loss": 5.5017,
      "mean_token_accuracy": 0.1620345413684845,
      "num_tokens": 28285983.0,
      "step": 13625
    },
    {
      "entropy": 5.765921354293823,
      "epoch": 1.2363933236574747,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048531999350105037,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15489896833896638,
      "num_tokens": 28296178.0,
      "step": 13630
    },
    {
      "entropy": 5.797476768493652,
      "epoch": 1.2368468795355587,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048530849536290537,
      "loss": 5.5598,
      "mean_token_accuracy": 0.15178377255797387,
      "num_tokens": 28305741.0,
      "step": 13635
    },
    {
      "entropy": 5.815008878707886,
      "epoch": 1.237300435413643,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004852969928755271,
      "loss": 5.5005,
      "mean_token_accuracy": 0.16100683957338333,
      "num_tokens": 28315607.0,
      "step": 13640
    },
    {
      "entropy": 5.777803516387939,
      "epoch": 1.237753991291727,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048528548603915344,
      "loss": 5.5329,
      "mean_token_accuracy": 0.16467561423778534,
      "num_tokens": 28327349.0,
      "step": 13645
    },
    {
      "entropy": 5.757795095443726,
      "epoch": 1.2382075471698113,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048527397485402236,
      "loss": 5.5191,
      "mean_token_accuracy": 0.15922290533781053,
      "num_tokens": 28339949.0,
      "step": 13650
    },
    {
      "entropy": 5.75310492515564,
      "epoch": 1.2386611030478956,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048526245932037186,
      "loss": 5.549,
      "mean_token_accuracy": 0.1637611135840416,
      "num_tokens": 28351340.0,
      "step": 13655
    },
    {
      "entropy": 5.674230670928955,
      "epoch": 1.2391146589259796,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048525093943844013,
      "loss": 5.4067,
      "mean_token_accuracy": 0.17018937319517136,
      "num_tokens": 28360929.0,
      "step": 13660
    },
    {
      "entropy": 5.745949745178223,
      "epoch": 1.2395682148040639,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004852394152084654,
      "loss": 5.5032,
      "mean_token_accuracy": 0.1667313665151596,
      "num_tokens": 28370477.0,
      "step": 13665
    },
    {
      "entropy": 5.7320106506347654,
      "epoch": 1.240021770682148,
      "grad_norm": 2.359375,
      "learning_rate": 0.000485227886630686,
      "loss": 5.4295,
      "mean_token_accuracy": 0.17114804238080977,
      "num_tokens": 28380876.0,
      "step": 13670
    },
    {
      "entropy": 5.711650419235229,
      "epoch": 1.2404753265602322,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004852163537053403,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16214274019002914,
      "num_tokens": 28390685.0,
      "step": 13675
    },
    {
      "entropy": 5.717212438583374,
      "epoch": 1.2409288824383164,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004852048164326669,
      "loss": 5.5193,
      "mean_token_accuracy": 0.15471189916133882,
      "num_tokens": 28400236.0,
      "step": 13680
    },
    {
      "entropy": 5.774027919769287,
      "epoch": 1.2413824383164005,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004851932748129043,
      "loss": 5.5572,
      "mean_token_accuracy": 0.16432497501373292,
      "num_tokens": 28410353.0,
      "step": 13685
    },
    {
      "entropy": 5.810084247589112,
      "epoch": 1.2418359941944848,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004851817288462912,
      "loss": 5.4913,
      "mean_token_accuracy": 0.15970743894577027,
      "num_tokens": 28420241.0,
      "step": 13690
    },
    {
      "entropy": 5.737703466415406,
      "epoch": 1.242289550072569,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004851701785330663,
      "loss": 5.5496,
      "mean_token_accuracy": 0.15706902593374253,
      "num_tokens": 28430610.0,
      "step": 13695
    },
    {
      "entropy": 5.737661981582642,
      "epoch": 1.242743105950653,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004851586238734687,
      "loss": 5.4898,
      "mean_token_accuracy": 0.16705117374658585,
      "num_tokens": 28440026.0,
      "step": 13700
    },
    {
      "entropy": 5.7917712211608885,
      "epoch": 1.2431966618287373,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004851470648677371,
      "loss": 5.5735,
      "mean_token_accuracy": 0.15307271629571914,
      "num_tokens": 28450707.0,
      "step": 13705
    },
    {
      "entropy": 5.776009845733642,
      "epoch": 1.2436502177068216,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004851355015161107,
      "loss": 5.5266,
      "mean_token_accuracy": 0.15916661471128463,
      "num_tokens": 28462378.0,
      "step": 13710
    },
    {
      "entropy": 5.808631324768067,
      "epoch": 1.2441037735849056,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004851239338188286,
      "loss": 5.5864,
      "mean_token_accuracy": 0.15191916823387147,
      "num_tokens": 28473084.0,
      "step": 13715
    },
    {
      "entropy": 5.757274961471557,
      "epoch": 1.24455732946299,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048511236177612994,
      "loss": 5.5126,
      "mean_token_accuracy": 0.16168709695339203,
      "num_tokens": 28482767.0,
      "step": 13720
    },
    {
      "entropy": 5.777504253387451,
      "epoch": 1.245010885341074,
      "grad_norm": 2.25,
      "learning_rate": 0.00048510078538825413,
      "loss": 5.5178,
      "mean_token_accuracy": 0.16122613251209258,
      "num_tokens": 28493274.0,
      "step": 13725
    },
    {
      "entropy": 5.812512350082398,
      "epoch": 1.2454644412191582,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048508920465544056,
      "loss": 5.5109,
      "mean_token_accuracy": 0.16319879442453383,
      "num_tokens": 28503823.0,
      "step": 13730
    },
    {
      "entropy": 5.703599357604981,
      "epoch": 1.2459179970972425,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004850776195779286,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1653244748711586,
      "num_tokens": 28514209.0,
      "step": 13735
    },
    {
      "entropy": 5.808743286132812,
      "epoch": 1.2463715529753265,
      "grad_norm": 3.640625,
      "learning_rate": 0.000485066030155958,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15484327375888823,
      "num_tokens": 28524665.0,
      "step": 13740
    },
    {
      "entropy": 5.75864052772522,
      "epoch": 1.2468251088534108,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004850544363897684,
      "loss": 5.4674,
      "mean_token_accuracy": 0.16477408707141877,
      "num_tokens": 28533639.0,
      "step": 13745
    },
    {
      "entropy": 5.672727012634278,
      "epoch": 1.2472786647314948,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048504283827959947,
      "loss": 5.5043,
      "mean_token_accuracy": 0.16229662001132966,
      "num_tokens": 28544469.0,
      "step": 13750
    },
    {
      "entropy": 5.846871757507325,
      "epoch": 1.2477322206095791,
      "grad_norm": 14.375,
      "learning_rate": 0.0004850312358256912,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1657694935798645,
      "num_tokens": 28554637.0,
      "step": 13755
    },
    {
      "entropy": 5.767649984359741,
      "epoch": 1.2481857764876634,
      "grad_norm": 3.0,
      "learning_rate": 0.0004850196290282834,
      "loss": 5.5047,
      "mean_token_accuracy": 0.1675608739256859,
      "num_tokens": 28564752.0,
      "step": 13760
    },
    {
      "entropy": 5.7551336765289305,
      "epoch": 1.2486393323657474,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048500801788761613,
      "loss": 5.6398,
      "mean_token_accuracy": 0.16060602962970733,
      "num_tokens": 28575007.0,
      "step": 13765
    },
    {
      "entropy": 5.84649338722229,
      "epoch": 1.2490928882438317,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004849964024039296,
      "loss": 5.5332,
      "mean_token_accuracy": 0.15880368351936341,
      "num_tokens": 28584907.0,
      "step": 13770
    },
    {
      "entropy": 5.87779598236084,
      "epoch": 1.2495464441219157,
      "grad_norm": 4.125,
      "learning_rate": 0.00048498478257746387,
      "loss": 5.6024,
      "mean_token_accuracy": 0.16832903772592545,
      "num_tokens": 28594244.0,
      "step": 13775
    },
    {
      "entropy": 5.803061485290527,
      "epoch": 1.25,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048497315840845934,
      "loss": 5.5566,
      "mean_token_accuracy": 0.16385678797960282,
      "num_tokens": 28603816.0,
      "step": 13780
    },
    {
      "entropy": 5.836783361434937,
      "epoch": 1.2504535558780843,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004849615298971565,
      "loss": 5.6002,
      "mean_token_accuracy": 0.15697139054536818,
      "num_tokens": 28613067.0,
      "step": 13785
    },
    {
      "entropy": 5.8058837890625,
      "epoch": 1.2509071117561683,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048494989704379556,
      "loss": 5.534,
      "mean_token_accuracy": 0.1632150173187256,
      "num_tokens": 28623752.0,
      "step": 13790
    },
    {
      "entropy": 5.737762928009033,
      "epoch": 1.2513606676342526,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004849382598486173,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1635163515806198,
      "num_tokens": 28633190.0,
      "step": 13795
    },
    {
      "entropy": 5.880742597579956,
      "epoch": 1.2518142235123366,
      "grad_norm": 2.25,
      "learning_rate": 0.00048492661831186243,
      "loss": 5.5979,
      "mean_token_accuracy": 0.15681124329566956,
      "num_tokens": 28644567.0,
      "step": 13800
    },
    {
      "entropy": 5.790067291259765,
      "epoch": 1.2522677793904209,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004849149724337715,
      "loss": 5.5886,
      "mean_token_accuracy": 0.15538534969091417,
      "num_tokens": 28654785.0,
      "step": 13805
    },
    {
      "entropy": 5.715815687179566,
      "epoch": 1.2527213352685052,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048490332221458543,
      "loss": 5.4639,
      "mean_token_accuracy": 0.16390922367572786,
      "num_tokens": 28665286.0,
      "step": 13810
    },
    {
      "entropy": 5.823144960403442,
      "epoch": 1.2531748911465892,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004848916676545452,
      "loss": 5.6074,
      "mean_token_accuracy": 0.16267772167921066,
      "num_tokens": 28675830.0,
      "step": 13815
    },
    {
      "entropy": 5.85063362121582,
      "epoch": 1.2536284470246735,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048488000875389174,
      "loss": 5.6478,
      "mean_token_accuracy": 0.1536603882908821,
      "num_tokens": 28686640.0,
      "step": 13820
    },
    {
      "entropy": 5.852750968933106,
      "epoch": 1.2540820029027575,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004848683455128663,
      "loss": 5.5054,
      "mean_token_accuracy": 0.16397723704576492,
      "num_tokens": 28697677.0,
      "step": 13825
    },
    {
      "entropy": 5.819412517547607,
      "epoch": 1.2545355587808418,
      "grad_norm": 3.9375,
      "learning_rate": 0.00048485667793170994,
      "loss": 5.5406,
      "mean_token_accuracy": 0.15668074935674667,
      "num_tokens": 28708831.0,
      "step": 13830
    },
    {
      "entropy": 5.748212909698486,
      "epoch": 1.254989114658926,
      "grad_norm": 3.09375,
      "learning_rate": 0.000484845006010664,
      "loss": 5.5647,
      "mean_token_accuracy": 0.16061486005783082,
      "num_tokens": 28718959.0,
      "step": 13835
    },
    {
      "entropy": 5.785844564437866,
      "epoch": 1.25544267053701,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004848333297499699,
      "loss": 5.5014,
      "mean_token_accuracy": 0.16540126800537108,
      "num_tokens": 28728719.0,
      "step": 13840
    },
    {
      "entropy": 5.868610525131226,
      "epoch": 1.2558962264150944,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048482164914986903,
      "loss": 5.6658,
      "mean_token_accuracy": 0.15237481892108917,
      "num_tokens": 28740032.0,
      "step": 13845
    },
    {
      "entropy": 5.772649765014648,
      "epoch": 1.2563497822931784,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048480996421060307,
      "loss": 5.5103,
      "mean_token_accuracy": 0.1705942541360855,
      "num_tokens": 28749374.0,
      "step": 13850
    },
    {
      "entropy": 5.710573959350586,
      "epoch": 1.2568033381712627,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048479827493241347,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1599291205406189,
      "num_tokens": 28760038.0,
      "step": 13855
    },
    {
      "entropy": 5.782832956314087,
      "epoch": 1.257256894049347,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048478658131554216,
      "loss": 5.5281,
      "mean_token_accuracy": 0.1633973613381386,
      "num_tokens": 28770247.0,
      "step": 13860
    },
    {
      "entropy": 5.798686933517456,
      "epoch": 1.2577104499274312,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004847748833602308,
      "loss": 5.4989,
      "mean_token_accuracy": 0.1631857141852379,
      "num_tokens": 28780251.0,
      "step": 13865
    },
    {
      "entropy": 5.7163026332855225,
      "epoch": 1.2581640058055152,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004847631810667215,
      "loss": 5.497,
      "mean_token_accuracy": 0.16497850120067598,
      "num_tokens": 28790829.0,
      "step": 13870
    },
    {
      "entropy": 5.665645503997803,
      "epoch": 1.2586175616835993,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004847514744352561,
      "loss": 5.4233,
      "mean_token_accuracy": 0.16679908484220504,
      "num_tokens": 28801036.0,
      "step": 13875
    },
    {
      "entropy": 5.682470417022705,
      "epoch": 1.2590711175616836,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004847397634660768,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16574569195508956,
      "num_tokens": 28811013.0,
      "step": 13880
    },
    {
      "entropy": 5.7319908618927,
      "epoch": 1.2595246734397678,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048472804815942575,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1625072777271271,
      "num_tokens": 28821367.0,
      "step": 13885
    },
    {
      "entropy": 5.741423416137695,
      "epoch": 1.259978229317852,
      "grad_norm": 9.4375,
      "learning_rate": 0.00048471632851554517,
      "loss": 5.5037,
      "mean_token_accuracy": 0.16509196162223816,
      "num_tokens": 28831077.0,
      "step": 13890
    },
    {
      "entropy": 5.819316148757935,
      "epoch": 1.2604317851959361,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004847046045346775,
      "loss": 5.545,
      "mean_token_accuracy": 0.15808188319206237,
      "num_tokens": 28841490.0,
      "step": 13895
    },
    {
      "entropy": 5.813286972045899,
      "epoch": 1.2608853410740204,
      "grad_norm": 3.3125,
      "learning_rate": 0.00048469287621706516,
      "loss": 5.5364,
      "mean_token_accuracy": 0.15896734446287156,
      "num_tokens": 28852510.0,
      "step": 13900
    },
    {
      "entropy": 5.764308786392212,
      "epoch": 1.2613388969521044,
      "grad_norm": 4.875,
      "learning_rate": 0.0004846811435629507,
      "loss": 5.5323,
      "mean_token_accuracy": 0.16363883316516875,
      "num_tokens": 28863388.0,
      "step": 13905
    },
    {
      "entropy": 5.6692163944244385,
      "epoch": 1.2617924528301887,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004846694065725768,
      "loss": 5.4839,
      "mean_token_accuracy": 0.16129184663295745,
      "num_tokens": 28873798.0,
      "step": 13910
    },
    {
      "entropy": 5.7175966739654545,
      "epoch": 1.262246008708273,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004846576652461861,
      "loss": 5.5138,
      "mean_token_accuracy": 0.15792021304368972,
      "num_tokens": 28883482.0,
      "step": 13915
    },
    {
      "entropy": 5.699329280853272,
      "epoch": 1.262699564586357,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048464591958402147,
      "loss": 5.4639,
      "mean_token_accuracy": 0.16810132414102555,
      "num_tokens": 28893573.0,
      "step": 13920
    },
    {
      "entropy": 5.819882154464722,
      "epoch": 1.2631531204644413,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048463416958632573,
      "loss": 5.6038,
      "mean_token_accuracy": 0.15934572666883468,
      "num_tokens": 28904064.0,
      "step": 13925
    },
    {
      "entropy": 5.827739667892456,
      "epoch": 1.2636066763425253,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048462241525334206,
      "loss": 5.5496,
      "mean_token_accuracy": 0.16146028637886048,
      "num_tokens": 28915329.0,
      "step": 13930
    },
    {
      "entropy": 5.734217405319214,
      "epoch": 1.2640602322206096,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004846106565853133,
      "loss": 5.6193,
      "mean_token_accuracy": 0.15644650757312775,
      "num_tokens": 28925228.0,
      "step": 13935
    },
    {
      "entropy": 5.759213638305664,
      "epoch": 1.2645137880986939,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048459889358248284,
      "loss": 5.4719,
      "mean_token_accuracy": 0.16382651329040526,
      "num_tokens": 28935891.0,
      "step": 13940
    },
    {
      "entropy": 5.745067834854126,
      "epoch": 1.264967343976778,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048458712624509383,
      "loss": 5.4612,
      "mean_token_accuracy": 0.16184199303388597,
      "num_tokens": 28946372.0,
      "step": 13945
    },
    {
      "entropy": 5.764742708206176,
      "epoch": 1.2654208998548622,
      "grad_norm": 3.578125,
      "learning_rate": 0.00048457535457338964,
      "loss": 5.5291,
      "mean_token_accuracy": 0.15864042192697525,
      "num_tokens": 28957874.0,
      "step": 13950
    },
    {
      "entropy": 5.811583757400513,
      "epoch": 1.2658744557329462,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004845635785676137,
      "loss": 5.55,
      "mean_token_accuracy": 0.15318827331066132,
      "num_tokens": 28968023.0,
      "step": 13955
    },
    {
      "entropy": 5.781111383438111,
      "epoch": 1.2663280116110305,
      "grad_norm": 4.59375,
      "learning_rate": 0.00048455179822800957,
      "loss": 5.4321,
      "mean_token_accuracy": 0.16802369952201843,
      "num_tokens": 28977744.0,
      "step": 13960
    },
    {
      "entropy": 5.762021875381469,
      "epoch": 1.2667815674891147,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004845400135548208,
      "loss": 5.6904,
      "mean_token_accuracy": 0.15434461534023286,
      "num_tokens": 28988875.0,
      "step": 13965
    },
    {
      "entropy": 5.763775062561035,
      "epoch": 1.2672351233671988,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004845282245482912,
      "loss": 5.4892,
      "mean_token_accuracy": 0.16463785916566848,
      "num_tokens": 28998562.0,
      "step": 13970
    },
    {
      "entropy": 5.786831378936768,
      "epoch": 1.267688679245283,
      "grad_norm": 2.375,
      "learning_rate": 0.0004845164312086646,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1669171079993248,
      "num_tokens": 29008783.0,
      "step": 13975
    },
    {
      "entropy": 5.665074110031128,
      "epoch": 1.268142235123367,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004845046335361847,
      "loss": 5.4245,
      "mean_token_accuracy": 0.16563522815704346,
      "num_tokens": 29019303.0,
      "step": 13980
    },
    {
      "entropy": 5.812286472320556,
      "epoch": 1.2685957910014514,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048449283153109566,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15621469765901566,
      "num_tokens": 29029843.0,
      "step": 13985
    },
    {
      "entropy": 5.850940322875976,
      "epoch": 1.2690493468795356,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004844810251936413,
      "loss": 5.5946,
      "mean_token_accuracy": 0.15403727889060975,
      "num_tokens": 29040603.0,
      "step": 13990
    },
    {
      "entropy": 5.744808101654053,
      "epoch": 1.2695029027576197,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004844692145240661,
      "loss": 5.4912,
      "mean_token_accuracy": 0.16592780947685243,
      "num_tokens": 29050599.0,
      "step": 13995
    },
    {
      "entropy": 5.773719358444214,
      "epoch": 1.269956458635704,
      "grad_norm": 2.875,
      "learning_rate": 0.0004844573995226142,
      "loss": 5.5646,
      "mean_token_accuracy": 0.15626720786094667,
      "num_tokens": 29061292.0,
      "step": 14000
    },
    {
      "entropy": 5.7530913829803465,
      "epoch": 1.270410014513788,
      "grad_norm": 2.375,
      "learning_rate": 0.00048444558018952984,
      "loss": 5.478,
      "mean_token_accuracy": 0.16829051822423935,
      "num_tokens": 29071779.0,
      "step": 14005
    },
    {
      "entropy": 5.740499353408813,
      "epoch": 1.2708635703918723,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048443375652505754,
      "loss": 5.4483,
      "mean_token_accuracy": 0.16108171492815018,
      "num_tokens": 29082073.0,
      "step": 14010
    },
    {
      "entropy": 5.753108739852905,
      "epoch": 1.2713171262699565,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004844219285294418,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1630928784608841,
      "num_tokens": 29092032.0,
      "step": 14015
    },
    {
      "entropy": 5.699787044525147,
      "epoch": 1.2717706821480406,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048441009620292714,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1686987727880478,
      "num_tokens": 29102936.0,
      "step": 14020
    },
    {
      "entropy": 5.751429605484009,
      "epoch": 1.2722242380261248,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004843982595457584,
      "loss": 5.498,
      "mean_token_accuracy": 0.16789738684892655,
      "num_tokens": 29113385.0,
      "step": 14025
    },
    {
      "entropy": 5.692891311645508,
      "epoch": 1.2726777939042089,
      "grad_norm": 3.125,
      "learning_rate": 0.00048438641855818026,
      "loss": 5.5481,
      "mean_token_accuracy": 0.1572081923484802,
      "num_tokens": 29125075.0,
      "step": 14030
    },
    {
      "entropy": 5.7562675952911375,
      "epoch": 1.2731313497822931,
      "grad_norm": 2.625,
      "learning_rate": 0.0004843745732404376,
      "loss": 5.4797,
      "mean_token_accuracy": 0.16220899075269699,
      "num_tokens": 29135098.0,
      "step": 14035
    },
    {
      "entropy": 5.815700340270996,
      "epoch": 1.2735849056603774,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004843627235927755,
      "loss": 5.5703,
      "mean_token_accuracy": 0.1634407952427864,
      "num_tokens": 29145577.0,
      "step": 14040
    },
    {
      "entropy": 5.743338394165039,
      "epoch": 1.2740384615384617,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004843508696154388,
      "loss": 5.5235,
      "mean_token_accuracy": 0.16221195235848426,
      "num_tokens": 29155905.0,
      "step": 14045
    },
    {
      "entropy": 5.680938863754273,
      "epoch": 1.2744920174165457,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048433901130867283,
      "loss": 5.4291,
      "mean_token_accuracy": 0.1655236378312111,
      "num_tokens": 29164926.0,
      "step": 14050
    },
    {
      "entropy": 5.71426420211792,
      "epoch": 1.2749455732946298,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004843271486727228,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16646048426628113,
      "num_tokens": 29175050.0,
      "step": 14055
    },
    {
      "entropy": 5.7006443500518795,
      "epoch": 1.275399129172714,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048431528170783395,
      "loss": 5.554,
      "mean_token_accuracy": 0.15971005260944365,
      "num_tokens": 29186255.0,
      "step": 14060
    },
    {
      "entropy": 5.794388341903686,
      "epoch": 1.2758526850507983,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004843034104142517,
      "loss": 5.5299,
      "mean_token_accuracy": 0.17041310518980027,
      "num_tokens": 29195639.0,
      "step": 14065
    },
    {
      "entropy": 5.827117013931274,
      "epoch": 1.2763062409288826,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004842915347922216,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1548395574092865,
      "num_tokens": 29206482.0,
      "step": 14070
    },
    {
      "entropy": 5.791819858551025,
      "epoch": 1.2767597968069666,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004842796548419893,
      "loss": 5.5209,
      "mean_token_accuracy": 0.1608733505010605,
      "num_tokens": 29217433.0,
      "step": 14075
    },
    {
      "entropy": 5.765908861160279,
      "epoch": 1.2772133526850509,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048426777056380035,
      "loss": 5.561,
      "mean_token_accuracy": 0.15637293457984924,
      "num_tokens": 29226193.0,
      "step": 14080
    },
    {
      "entropy": 5.6839453220367435,
      "epoch": 1.277666908563135,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004842558819579006,
      "loss": 5.5099,
      "mean_token_accuracy": 0.1635231629014015,
      "num_tokens": 29237061.0,
      "step": 14085
    },
    {
      "entropy": 5.706640005111694,
      "epoch": 1.2781204644412192,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004842439890245358,
      "loss": 5.4903,
      "mean_token_accuracy": 0.16808591932058334,
      "num_tokens": 29247689.0,
      "step": 14090
    },
    {
      "entropy": 5.746940040588379,
      "epoch": 1.2785740203193035,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048423209176395204,
      "loss": 5.472,
      "mean_token_accuracy": 0.16666798889636994,
      "num_tokens": 29258235.0,
      "step": 14095
    },
    {
      "entropy": 5.7619181156158445,
      "epoch": 1.2790275761973875,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004842201901763954,
      "loss": 5.4753,
      "mean_token_accuracy": 0.16424423456192017,
      "num_tokens": 29268884.0,
      "step": 14100
    },
    {
      "entropy": 5.732350254058838,
      "epoch": 1.2794811320754718,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004842082842621118,
      "loss": 5.602,
      "mean_token_accuracy": 0.1559084475040436,
      "num_tokens": 29280484.0,
      "step": 14105
    },
    {
      "entropy": 5.786956357955932,
      "epoch": 1.2799346879535558,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004841963740213476,
      "loss": 5.5621,
      "mean_token_accuracy": 0.16354743540287017,
      "num_tokens": 29291199.0,
      "step": 14110
    },
    {
      "entropy": 5.833890581130982,
      "epoch": 1.28038824383164,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004841844594543491,
      "loss": 5.5233,
      "mean_token_accuracy": 0.1619634747505188,
      "num_tokens": 29302007.0,
      "step": 14115
    },
    {
      "entropy": 5.731606578826904,
      "epoch": 1.2808417997097243,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004841725405613626,
      "loss": 5.4992,
      "mean_token_accuracy": 0.16947626024484636,
      "num_tokens": 29312127.0,
      "step": 14120
    },
    {
      "entropy": 5.757757043838501,
      "epoch": 1.2812953555878084,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004841606173426347,
      "loss": 5.637,
      "mean_token_accuracy": 0.15512577146291734,
      "num_tokens": 29323691.0,
      "step": 14125
    },
    {
      "entropy": 5.73169322013855,
      "epoch": 1.2817489114658926,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048414868979841197,
      "loss": 5.5294,
      "mean_token_accuracy": 0.1576601728796959,
      "num_tokens": 29335203.0,
      "step": 14130
    },
    {
      "entropy": 5.842589807510376,
      "epoch": 1.2822024673439767,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048413675792894105,
      "loss": 5.582,
      "mean_token_accuracy": 0.1653008684515953,
      "num_tokens": 29346422.0,
      "step": 14135
    },
    {
      "entropy": 5.756655597686768,
      "epoch": 1.282656023222061,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004841248217344686,
      "loss": 5.48,
      "mean_token_accuracy": 0.1627561554312706,
      "num_tokens": 29356084.0,
      "step": 14140
    },
    {
      "entropy": 5.671939563751221,
      "epoch": 1.2831095791001452,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004841128812152415,
      "loss": 5.5024,
      "mean_token_accuracy": 0.16343217939138413,
      "num_tokens": 29366093.0,
      "step": 14145
    },
    {
      "entropy": 5.74494047164917,
      "epoch": 1.2835631349782293,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004841009363715068,
      "loss": 5.5589,
      "mean_token_accuracy": 0.15755545124411582,
      "num_tokens": 29376621.0,
      "step": 14150
    },
    {
      "entropy": 5.816412115097046,
      "epoch": 1.2840166908563135,
      "grad_norm": 2.5,
      "learning_rate": 0.0004840889872035115,
      "loss": 5.6099,
      "mean_token_accuracy": 0.16073900759220122,
      "num_tokens": 29387119.0,
      "step": 14155
    },
    {
      "entropy": 5.760625123977661,
      "epoch": 1.2844702467343976,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004840770337115026,
      "loss": 5.521,
      "mean_token_accuracy": 0.1657382905483246,
      "num_tokens": 29396981.0,
      "step": 14160
    },
    {
      "entropy": 5.767019271850586,
      "epoch": 1.2849238026124818,
      "grad_norm": 2.875,
      "learning_rate": 0.00048406507589572747,
      "loss": 5.5074,
      "mean_token_accuracy": 0.16722526848316194,
      "num_tokens": 29406737.0,
      "step": 14165
    },
    {
      "entropy": 5.751604318618774,
      "epoch": 1.2853773584905661,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004840531137564331,
      "loss": 5.5121,
      "mean_token_accuracy": 0.15941444635391236,
      "num_tokens": 29417300.0,
      "step": 14170
    },
    {
      "entropy": 5.742380428314209,
      "epoch": 1.2858309143686502,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004840411472938672,
      "loss": 5.4359,
      "mean_token_accuracy": 0.17064938098192214,
      "num_tokens": 29426749.0,
      "step": 14175
    },
    {
      "entropy": 5.7362236976623535,
      "epoch": 1.2862844702467344,
      "grad_norm": 3.625,
      "learning_rate": 0.0004840291765082771,
      "loss": 5.4584,
      "mean_token_accuracy": 0.16377734392881393,
      "num_tokens": 29437264.0,
      "step": 14180
    },
    {
      "entropy": 5.695000839233399,
      "epoch": 1.2867380261248185,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048401720139991036,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1571073815226555,
      "num_tokens": 29448192.0,
      "step": 14185
    },
    {
      "entropy": 5.7379358291625975,
      "epoch": 1.2871915820029027,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004840052219690146,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1709623396396637,
      "num_tokens": 29457923.0,
      "step": 14190
    },
    {
      "entropy": 5.682750558853149,
      "epoch": 1.287645137880987,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004839932382158376,
      "loss": 5.4522,
      "mean_token_accuracy": 0.1727570042014122,
      "num_tokens": 29466914.0,
      "step": 14195
    },
    {
      "entropy": 5.709893465042114,
      "epoch": 1.288098693759071,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004839812501406272,
      "loss": 5.5451,
      "mean_token_accuracy": 0.16052595376968384,
      "num_tokens": 29477648.0,
      "step": 14200
    },
    {
      "entropy": 5.786252212524414,
      "epoch": 1.2885522496371553,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004839692577436313,
      "loss": 5.5229,
      "mean_token_accuracy": 0.1661274254322052,
      "num_tokens": 29487947.0,
      "step": 14205
    },
    {
      "entropy": 5.86531629562378,
      "epoch": 1.2890058055152394,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004839572610250978,
      "loss": 5.6048,
      "mean_token_accuracy": 0.157396100461483,
      "num_tokens": 29497667.0,
      "step": 14210
    },
    {
      "entropy": 5.705365562438965,
      "epoch": 1.2894593613933236,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048394525998527496,
      "loss": 5.4881,
      "mean_token_accuracy": 0.15912211537361146,
      "num_tokens": 29507065.0,
      "step": 14215
    },
    {
      "entropy": 5.732524156570435,
      "epoch": 1.2899129172714079,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004839332546244109,
      "loss": 5.6146,
      "mean_token_accuracy": 0.1581760197877884,
      "num_tokens": 29517752.0,
      "step": 14220
    },
    {
      "entropy": 5.7949902534484865,
      "epoch": 1.2903664731494922,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048392124494275385,
      "loss": 5.5076,
      "mean_token_accuracy": 0.15902452021837235,
      "num_tokens": 29527894.0,
      "step": 14225
    },
    {
      "entropy": 5.786681413650513,
      "epoch": 1.2908200290275762,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004839092309405523,
      "loss": 5.5529,
      "mean_token_accuracy": 0.16011998504400254,
      "num_tokens": 29538136.0,
      "step": 14230
    },
    {
      "entropy": 5.763975763320923,
      "epoch": 1.2912735849056602,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048389721261805457,
      "loss": 5.5067,
      "mean_token_accuracy": 0.16358981281518936,
      "num_tokens": 29549193.0,
      "step": 14235
    },
    {
      "entropy": 5.713531875610352,
      "epoch": 1.2917271407837445,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048388518997550926,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16910220682621002,
      "num_tokens": 29558889.0,
      "step": 14240
    },
    {
      "entropy": 5.780442714691162,
      "epoch": 1.2921806966618288,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048387316301316493,
      "loss": 5.545,
      "mean_token_accuracy": 0.15655036717653276,
      "num_tokens": 29569691.0,
      "step": 14245
    },
    {
      "entropy": 5.774052429199219,
      "epoch": 1.292634252539913,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004838611317312705,
      "loss": 5.5799,
      "mean_token_accuracy": 0.16386362314224243,
      "num_tokens": 29580459.0,
      "step": 14250
    },
    {
      "entropy": 5.75199670791626,
      "epoch": 1.293087808417997,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004838490961300745,
      "loss": 5.5185,
      "mean_token_accuracy": 0.16147345602512359,
      "num_tokens": 29590652.0,
      "step": 14255
    },
    {
      "entropy": 5.718763542175293,
      "epoch": 1.2935413642960814,
      "grad_norm": 9.8125,
      "learning_rate": 0.00048383705620982606,
      "loss": 5.4915,
      "mean_token_accuracy": 0.17403658032417296,
      "num_tokens": 29600353.0,
      "step": 14260
    },
    {
      "entropy": 5.734799385070801,
      "epoch": 1.2939949201741654,
      "grad_norm": 2.5,
      "learning_rate": 0.00048382501197077405,
      "loss": 5.5032,
      "mean_token_accuracy": 0.17103185951709748,
      "num_tokens": 29610561.0,
      "step": 14265
    },
    {
      "entropy": 5.711618423461914,
      "epoch": 1.2944484760522497,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004838129634131675,
      "loss": 5.5136,
      "mean_token_accuracy": 0.16385139226913453,
      "num_tokens": 29621741.0,
      "step": 14270
    },
    {
      "entropy": 5.717280578613281,
      "epoch": 1.294902031930334,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004838009105372558,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16221119910478593,
      "num_tokens": 29633097.0,
      "step": 14275
    },
    {
      "entropy": 5.756360816955566,
      "epoch": 1.295355587808418,
      "grad_norm": 5.0,
      "learning_rate": 0.00048378885334328793,
      "loss": 5.5488,
      "mean_token_accuracy": 0.1551951378583908,
      "num_tokens": 29644290.0,
      "step": 14280
    },
    {
      "entropy": 5.829109144210816,
      "epoch": 1.2958091436865022,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004837767918315135,
      "loss": 5.5948,
      "mean_token_accuracy": 0.1530178427696228,
      "num_tokens": 29654977.0,
      "step": 14285
    },
    {
      "entropy": 5.792234086990357,
      "epoch": 1.2962626995645863,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048376472600218174,
      "loss": 5.5763,
      "mean_token_accuracy": 0.15937286615371704,
      "num_tokens": 29665020.0,
      "step": 14290
    },
    {
      "entropy": 5.7204102039337155,
      "epoch": 1.2967162554426706,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004837526558555422,
      "loss": 5.5259,
      "mean_token_accuracy": 0.16547319591045379,
      "num_tokens": 29675098.0,
      "step": 14295
    },
    {
      "entropy": 5.732232332229614,
      "epoch": 1.2971698113207548,
      "grad_norm": 2.625,
      "learning_rate": 0.0004837405813918446,
      "loss": 5.4704,
      "mean_token_accuracy": 0.16832402348518372,
      "num_tokens": 29685450.0,
      "step": 14300
    },
    {
      "entropy": 5.711061954498291,
      "epoch": 1.2976233671988389,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048372850261133856,
      "loss": 5.466,
      "mean_token_accuracy": 0.1615486592054367,
      "num_tokens": 29694818.0,
      "step": 14305
    },
    {
      "entropy": 5.77941665649414,
      "epoch": 1.2980769230769231,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048371641951427393,
      "loss": 5.563,
      "mean_token_accuracy": 0.15773442536592483,
      "num_tokens": 29705573.0,
      "step": 14310
    },
    {
      "entropy": 5.787207555770874,
      "epoch": 1.2985304789550072,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048370433210090055,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1600997507572174,
      "num_tokens": 29716366.0,
      "step": 14315
    },
    {
      "entropy": 5.754968357086182,
      "epoch": 1.2989840348330914,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004836922403714683,
      "loss": 5.4934,
      "mean_token_accuracy": 0.16566208153963088,
      "num_tokens": 29727181.0,
      "step": 14320
    },
    {
      "entropy": 5.727057695388794,
      "epoch": 1.2994375907111757,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048368014432622746,
      "loss": 5.5002,
      "mean_token_accuracy": 0.16421906650066376,
      "num_tokens": 29736960.0,
      "step": 14325
    },
    {
      "entropy": 5.747868251800537,
      "epoch": 1.2998911465892597,
      "grad_norm": 5.125,
      "learning_rate": 0.000483668043965428,
      "loss": 5.4001,
      "mean_token_accuracy": 0.17528508305549623,
      "num_tokens": 29746354.0,
      "step": 14330
    },
    {
      "entropy": 5.727923059463501,
      "epoch": 1.300344702467344,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004836559392893203,
      "loss": 5.5027,
      "mean_token_accuracy": 0.1624224916100502,
      "num_tokens": 29756627.0,
      "step": 14335
    },
    {
      "entropy": 5.720639181137085,
      "epoch": 1.300798258345428,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004836438302981545,
      "loss": 5.4971,
      "mean_token_accuracy": 0.16663634032011032,
      "num_tokens": 29766684.0,
      "step": 14340
    },
    {
      "entropy": 5.715429973602295,
      "epoch": 1.3012518142235123,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004836317169921811,
      "loss": 5.5288,
      "mean_token_accuracy": 0.16115170568227768,
      "num_tokens": 29776556.0,
      "step": 14345
    },
    {
      "entropy": 5.833919191360474,
      "epoch": 1.3017053701015966,
      "grad_norm": 3.921875,
      "learning_rate": 0.00048361959937165075,
      "loss": 5.6222,
      "mean_token_accuracy": 0.1591704621911049,
      "num_tokens": 29786896.0,
      "step": 14350
    },
    {
      "entropy": 5.776287508010864,
      "epoch": 1.3021589259796806,
      "grad_norm": 3.75,
      "learning_rate": 0.0004836074774368138,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1661750629544258,
      "num_tokens": 29797332.0,
      "step": 14355
    },
    {
      "entropy": 5.780146074295044,
      "epoch": 1.302612481857765,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048359535118792116,
      "loss": 5.5251,
      "mean_token_accuracy": 0.16391671746969222,
      "num_tokens": 29806428.0,
      "step": 14360
    },
    {
      "entropy": 5.736402797698974,
      "epoch": 1.303066037735849,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004835832206252235,
      "loss": 5.5192,
      "mean_token_accuracy": 0.1575208842754364,
      "num_tokens": 29816794.0,
      "step": 14365
    },
    {
      "entropy": 5.8067232131958,
      "epoch": 1.3035195936139332,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004835710857489716,
      "loss": 5.552,
      "mean_token_accuracy": 0.1623404696583748,
      "num_tokens": 29827077.0,
      "step": 14370
    },
    {
      "entropy": 5.709520292282105,
      "epoch": 1.3039731494920175,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004835589465594165,
      "loss": 5.5019,
      "mean_token_accuracy": 0.16448380649089814,
      "num_tokens": 29837868.0,
      "step": 14375
    },
    {
      "entropy": 5.741806554794311,
      "epoch": 1.3044267053701015,
      "grad_norm": 3.9375,
      "learning_rate": 0.00048354680305680926,
      "loss": 5.5652,
      "mean_token_accuracy": 0.16395579576492308,
      "num_tokens": 29846929.0,
      "step": 14380
    },
    {
      "entropy": 5.726155853271484,
      "epoch": 1.3048802612481858,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048353465524140105,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16291254609823227,
      "num_tokens": 29857905.0,
      "step": 14385
    },
    {
      "entropy": 5.774167585372925,
      "epoch": 1.3053338171262698,
      "grad_norm": 5.59375,
      "learning_rate": 0.000483522503113443,
      "loss": 5.5309,
      "mean_token_accuracy": 0.1612105742096901,
      "num_tokens": 29868230.0,
      "step": 14390
    },
    {
      "entropy": 5.779612398147583,
      "epoch": 1.305787373004354,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048351034667318634,
      "loss": 5.4957,
      "mean_token_accuracy": 0.16405465453863144,
      "num_tokens": 29879224.0,
      "step": 14395
    },
    {
      "entropy": 5.82379903793335,
      "epoch": 1.3062409288824384,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004834981859208827,
      "loss": 5.6308,
      "mean_token_accuracy": 0.15578243434429168,
      "num_tokens": 29890416.0,
      "step": 14400
    },
    {
      "entropy": 5.7576545715332035,
      "epoch": 1.3066944847605226,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048348602085678346,
      "loss": 5.5606,
      "mean_token_accuracy": 0.15813945829868317,
      "num_tokens": 29901064.0,
      "step": 14405
    },
    {
      "entropy": 5.788601016998291,
      "epoch": 1.3071480406386067,
      "grad_norm": 3.609375,
      "learning_rate": 0.00048347385148114004,
      "loss": 5.6219,
      "mean_token_accuracy": 0.16267191171646117,
      "num_tokens": 29911084.0,
      "step": 14410
    },
    {
      "entropy": 5.710220861434936,
      "epoch": 1.3076015965166907,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048346167779420436,
      "loss": 5.4535,
      "mean_token_accuracy": 0.167217455804348,
      "num_tokens": 29921104.0,
      "step": 14415
    },
    {
      "entropy": 5.7539019107818605,
      "epoch": 1.308055152394775,
      "grad_norm": 2.25,
      "learning_rate": 0.00048344949979622806,
      "loss": 5.5012,
      "mean_token_accuracy": 0.16407532542943953,
      "num_tokens": 29932014.0,
      "step": 14420
    },
    {
      "entropy": 5.7296953201293945,
      "epoch": 1.3085087082728593,
      "grad_norm": 5.0,
      "learning_rate": 0.00048343731748746294,
      "loss": 5.5005,
      "mean_token_accuracy": 0.1659050390124321,
      "num_tokens": 29941707.0,
      "step": 14425
    },
    {
      "entropy": 5.701277160644532,
      "epoch": 1.3089622641509435,
      "grad_norm": 2.75,
      "learning_rate": 0.000483425130868161,
      "loss": 5.5031,
      "mean_token_accuracy": 0.15964508056640625,
      "num_tokens": 29952082.0,
      "step": 14430
    },
    {
      "entropy": 5.707495403289795,
      "epoch": 1.3094158200290276,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004834129399385742,
      "loss": 5.3692,
      "mean_token_accuracy": 0.17637512236833572,
      "num_tokens": 29961744.0,
      "step": 14435
    },
    {
      "entropy": 5.7725728988647464,
      "epoch": 1.3098693759071118,
      "grad_norm": 2.75,
      "learning_rate": 0.0004834007446989548,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1668949618935585,
      "num_tokens": 29972611.0,
      "step": 14440
    },
    {
      "entropy": 5.725486850738525,
      "epoch": 1.3103229317851959,
      "grad_norm": 5.03125,
      "learning_rate": 0.00048338854514955474,
      "loss": 5.4294,
      "mean_token_accuracy": 0.17146183103322982,
      "num_tokens": 29982660.0,
      "step": 14445
    },
    {
      "entropy": 5.71315279006958,
      "epoch": 1.3107764876632801,
      "grad_norm": 3.96875,
      "learning_rate": 0.00048337634129062655,
      "loss": 5.5222,
      "mean_token_accuracy": 0.15204458832740783,
      "num_tokens": 29993349.0,
      "step": 14450
    },
    {
      "entropy": 5.766847848892212,
      "epoch": 1.3112300435413644,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048336413312242255,
      "loss": 5.6105,
      "mean_token_accuracy": 0.1561606392264366,
      "num_tokens": 30004474.0,
      "step": 14455
    },
    {
      "entropy": 5.827823352813721,
      "epoch": 1.3116835994194485,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048335192064519514,
      "loss": 5.5705,
      "mean_token_accuracy": 0.1668871149420738,
      "num_tokens": 30015189.0,
      "step": 14460
    },
    {
      "entropy": 5.817894220352173,
      "epoch": 1.3121371552975327,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048333970385919695,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16282259970903395,
      "num_tokens": 30025491.0,
      "step": 14465
    },
    {
      "entropy": 5.720710897445679,
      "epoch": 1.3125907111756168,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048332748276468053,
      "loss": 5.5008,
      "mean_token_accuracy": 0.16462895274162292,
      "num_tokens": 30034683.0,
      "step": 14470
    },
    {
      "entropy": 5.719644165039062,
      "epoch": 1.313044267053701,
      "grad_norm": 2.875,
      "learning_rate": 0.0004833152573618987,
      "loss": 5.4847,
      "mean_token_accuracy": 0.16472677290439605,
      "num_tokens": 30044135.0,
      "step": 14475
    },
    {
      "entropy": 5.803503847122192,
      "epoch": 1.3134978229317853,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048330302765110436,
      "loss": 5.5879,
      "mean_token_accuracy": 0.16193737834692,
      "num_tokens": 30054520.0,
      "step": 14480
    },
    {
      "entropy": 5.813305854797363,
      "epoch": 1.3139513788098693,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048329079363255024,
      "loss": 5.5722,
      "mean_token_accuracy": 0.15950210094451905,
      "num_tokens": 30065439.0,
      "step": 14485
    },
    {
      "entropy": 5.789185810089111,
      "epoch": 1.3144049346879536,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048327855530648934,
      "loss": 5.5343,
      "mean_token_accuracy": 0.16219229996204376,
      "num_tokens": 30075362.0,
      "step": 14490
    },
    {
      "entropy": 5.765440130233765,
      "epoch": 1.3148584905660377,
      "grad_norm": 2.984375,
      "learning_rate": 0.000483266312673175,
      "loss": 5.5054,
      "mean_token_accuracy": 0.16190118193626404,
      "num_tokens": 30084913.0,
      "step": 14495
    },
    {
      "entropy": 5.763190650939942,
      "epoch": 1.315312046444122,
      "grad_norm": 2.875,
      "learning_rate": 0.0004832540657328602,
      "loss": 5.5127,
      "mean_token_accuracy": 0.16513173878192902,
      "num_tokens": 30095704.0,
      "step": 14500
    },
    {
      "entropy": 5.7071905612945555,
      "epoch": 1.3157656023222062,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004832418144857982,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16373568326234816,
      "num_tokens": 30105981.0,
      "step": 14505
    },
    {
      "entropy": 5.726767349243164,
      "epoch": 1.3162191582002902,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048322955893224245,
      "loss": 5.522,
      "mean_token_accuracy": 0.15650906264781952,
      "num_tokens": 30115772.0,
      "step": 14510
    },
    {
      "entropy": 5.790813302993774,
      "epoch": 1.3166727140783745,
      "grad_norm": 4.0,
      "learning_rate": 0.00048321729907244626,
      "loss": 5.4941,
      "mean_token_accuracy": 0.16158971935510635,
      "num_tokens": 30127050.0,
      "step": 14515
    },
    {
      "entropy": 5.751594209671021,
      "epoch": 1.3171262699564585,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004832050349066634,
      "loss": 5.5292,
      "mean_token_accuracy": 0.16067265272140502,
      "num_tokens": 30138221.0,
      "step": 14520
    },
    {
      "entropy": 5.790085792541504,
      "epoch": 1.3175798258345428,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004831927664351473,
      "loss": 5.5867,
      "mean_token_accuracy": 0.16004825085401536,
      "num_tokens": 30150181.0,
      "step": 14525
    },
    {
      "entropy": 5.806090402603149,
      "epoch": 1.318033381712627,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004831804936581518,
      "loss": 5.5671,
      "mean_token_accuracy": 0.16037465929985045,
      "num_tokens": 30160521.0,
      "step": 14530
    },
    {
      "entropy": 5.812496709823608,
      "epoch": 1.3184869375907111,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004831682165759305,
      "loss": 5.5359,
      "mean_token_accuracy": 0.1598220095038414,
      "num_tokens": 30170611.0,
      "step": 14535
    },
    {
      "entropy": 5.809488248825073,
      "epoch": 1.3189404934687954,
      "grad_norm": 3.0,
      "learning_rate": 0.00048315593518873757,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16837014108896256,
      "num_tokens": 30179914.0,
      "step": 14540
    },
    {
      "entropy": 5.774171781539917,
      "epoch": 1.3193940493468794,
      "grad_norm": 2.375,
      "learning_rate": 0.0004831436494968268,
      "loss": 5.5036,
      "mean_token_accuracy": 0.1644914045929909,
      "num_tokens": 30191426.0,
      "step": 14545
    },
    {
      "entropy": 5.700647354125977,
      "epoch": 1.3198476052249637,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048313135950045234,
      "loss": 5.4737,
      "mean_token_accuracy": 0.15935910642147064,
      "num_tokens": 30202813.0,
      "step": 14550
    },
    {
      "entropy": 5.700052070617676,
      "epoch": 1.320301161103048,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048311906519986824,
      "loss": 5.418,
      "mean_token_accuracy": 0.16932462751865388,
      "num_tokens": 30213044.0,
      "step": 14555
    },
    {
      "entropy": 5.8162455558776855,
      "epoch": 1.320754716981132,
      "grad_norm": 2.25,
      "learning_rate": 0.000483106766595329,
      "loss": 5.5025,
      "mean_token_accuracy": 0.15728759318590163,
      "num_tokens": 30223419.0,
      "step": 14560
    },
    {
      "entropy": 5.808168506622314,
      "epoch": 1.3212082728592163,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004830944636870887,
      "loss": 5.533,
      "mean_token_accuracy": 0.16001826226711274,
      "num_tokens": 30233203.0,
      "step": 14565
    },
    {
      "entropy": 5.763615560531616,
      "epoch": 1.3216618287373003,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048308215647540183,
      "loss": 5.5458,
      "mean_token_accuracy": 0.15754529386758803,
      "num_tokens": 30244604.0,
      "step": 14570
    },
    {
      "entropy": 5.779519748687744,
      "epoch": 1.3221153846153846,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048306984496052297,
      "loss": 5.5997,
      "mean_token_accuracy": 0.16379113346338273,
      "num_tokens": 30255355.0,
      "step": 14575
    },
    {
      "entropy": 5.811548185348511,
      "epoch": 1.3225689404934688,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048305752914270665,
      "loss": 5.6277,
      "mean_token_accuracy": 0.153807170689106,
      "num_tokens": 30265665.0,
      "step": 14580
    },
    {
      "entropy": 5.797720813751221,
      "epoch": 1.3230224963715531,
      "grad_norm": 3.625,
      "learning_rate": 0.00048304520902220767,
      "loss": 5.573,
      "mean_token_accuracy": 0.16290334165096282,
      "num_tokens": 30274347.0,
      "step": 14585
    },
    {
      "entropy": 5.78583025932312,
      "epoch": 1.3234760522496372,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004830328845992808,
      "loss": 5.4209,
      "mean_token_accuracy": 0.16693383753299712,
      "num_tokens": 30284243.0,
      "step": 14590
    },
    {
      "entropy": 5.699343919754028,
      "epoch": 1.3239296081277212,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048302055587418074,
      "loss": 5.4351,
      "mean_token_accuracy": 0.16768361181020736,
      "num_tokens": 30294327.0,
      "step": 14595
    },
    {
      "entropy": 5.687645673751831,
      "epoch": 1.3243831640058055,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004830082228471626,
      "loss": 5.5526,
      "mean_token_accuracy": 0.16276278346776962,
      "num_tokens": 30304188.0,
      "step": 14600
    },
    {
      "entropy": 5.711098623275757,
      "epoch": 1.3248367198838897,
      "grad_norm": 4.1875,
      "learning_rate": 0.00048299588551848146,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16924587786197662,
      "num_tokens": 30314917.0,
      "step": 14605
    },
    {
      "entropy": 5.736079359054566,
      "epoch": 1.325290275761974,
      "grad_norm": 3.59375,
      "learning_rate": 0.00048298354388839237,
      "loss": 5.4659,
      "mean_token_accuracy": 0.16330193877220153,
      "num_tokens": 30325221.0,
      "step": 14610
    },
    {
      "entropy": 5.823500156402588,
      "epoch": 1.325743831640058,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004829711979571506,
      "loss": 5.5631,
      "mean_token_accuracy": 0.15871859192848206,
      "num_tokens": 30336196.0,
      "step": 14615
    },
    {
      "entropy": 5.711117267608643,
      "epoch": 1.3261973875181423,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004829588477250115,
      "loss": 5.3802,
      "mean_token_accuracy": 0.17468658089637756,
      "num_tokens": 30345546.0,
      "step": 14620
    },
    {
      "entropy": 5.715722370147705,
      "epoch": 1.3266509433962264,
      "grad_norm": 2.25,
      "learning_rate": 0.0004829464931922304,
      "loss": 5.5256,
      "mean_token_accuracy": 0.16414328813552856,
      "num_tokens": 30355672.0,
      "step": 14625
    },
    {
      "entropy": 5.778738212585449,
      "epoch": 1.3271044992743106,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048293413435906285,
      "loss": 5.5128,
      "mean_token_accuracy": 0.1599799320101738,
      "num_tokens": 30366233.0,
      "step": 14630
    },
    {
      "entropy": 5.800315237045288,
      "epoch": 1.327558055152395,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004829217712257644,
      "loss": 5.519,
      "mean_token_accuracy": 0.1618216708302498,
      "num_tokens": 30377366.0,
      "step": 14635
    },
    {
      "entropy": 5.746817684173584,
      "epoch": 1.328011611030479,
      "grad_norm": 3.125,
      "learning_rate": 0.00048290940379259073,
      "loss": 5.5535,
      "mean_token_accuracy": 0.15986550748348236,
      "num_tokens": 30388022.0,
      "step": 14640
    },
    {
      "entropy": 5.752737760543823,
      "epoch": 1.3284651669085632,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048289703205979756,
      "loss": 5.4508,
      "mean_token_accuracy": 0.15991674214601517,
      "num_tokens": 30399235.0,
      "step": 14645
    },
    {
      "entropy": 5.844228792190552,
      "epoch": 1.3289187227866472,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048288465602764084,
      "loss": 5.5811,
      "mean_token_accuracy": 0.1616254724562168,
      "num_tokens": 30409112.0,
      "step": 14650
    },
    {
      "entropy": 5.760185813903808,
      "epoch": 1.3293722786647315,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048287227569637645,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1637389838695526,
      "num_tokens": 30420202.0,
      "step": 14655
    },
    {
      "entropy": 5.758245611190796,
      "epoch": 1.3298258345428158,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048285989106626036,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1651548773050308,
      "num_tokens": 30431210.0,
      "step": 14660
    },
    {
      "entropy": 5.715674161911011,
      "epoch": 1.3302793904208998,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004828475021375489,
      "loss": 5.4568,
      "mean_token_accuracy": 0.1669271782040596,
      "num_tokens": 30441291.0,
      "step": 14665
    },
    {
      "entropy": 5.715033864974975,
      "epoch": 1.330732946298984,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004828351089104981,
      "loss": 5.4144,
      "mean_token_accuracy": 0.17231065928936004,
      "num_tokens": 30450285.0,
      "step": 14670
    },
    {
      "entropy": 5.65064845085144,
      "epoch": 1.3311865021770681,
      "grad_norm": 6.25,
      "learning_rate": 0.00048282271138536424,
      "loss": 5.5167,
      "mean_token_accuracy": 0.16380294114351274,
      "num_tokens": 30460311.0,
      "step": 14675
    },
    {
      "entropy": 5.799954557418824,
      "epoch": 1.3316400580551524,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004828103095624037,
      "loss": 5.5525,
      "mean_token_accuracy": 0.16289848536252977,
      "num_tokens": 30470188.0,
      "step": 14680
    },
    {
      "entropy": 5.793932008743286,
      "epoch": 1.3320936139332367,
      "grad_norm": 7.84375,
      "learning_rate": 0.000482797903441873,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1587819278240204,
      "num_tokens": 30481406.0,
      "step": 14685
    },
    {
      "entropy": 5.72804446220398,
      "epoch": 1.3325471698113207,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004827854930240288,
      "loss": 5.4744,
      "mean_token_accuracy": 0.1691707640886307,
      "num_tokens": 30492314.0,
      "step": 14690
    },
    {
      "entropy": 5.80890965461731,
      "epoch": 1.333000725689405,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004827730783091276,
      "loss": 5.6116,
      "mean_token_accuracy": 0.15468821972608565,
      "num_tokens": 30503273.0,
      "step": 14695
    },
    {
      "entropy": 5.830076074600219,
      "epoch": 1.333454281567489,
      "grad_norm": 2.875,
      "learning_rate": 0.00048276065929742616,
      "loss": 5.5047,
      "mean_token_accuracy": 0.16333384662866593,
      "num_tokens": 30513582.0,
      "step": 14700
    },
    {
      "entropy": 5.709561061859131,
      "epoch": 1.3339078374455733,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004827482359891814,
      "loss": 5.4335,
      "mean_token_accuracy": 0.17266145944595337,
      "num_tokens": 30523261.0,
      "step": 14705
    },
    {
      "entropy": 5.713970136642456,
      "epoch": 1.3343613933236576,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048273580838465016,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1610859990119934,
      "num_tokens": 30533774.0,
      "step": 14710
    },
    {
      "entropy": 5.753443670272827,
      "epoch": 1.3348149492017416,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004827233764840894,
      "loss": 5.5826,
      "mean_token_accuracy": 0.15997211784124374,
      "num_tokens": 30544580.0,
      "step": 14715
    },
    {
      "entropy": 5.7764476299285885,
      "epoch": 1.3352685050798259,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004827109402877564,
      "loss": 5.4422,
      "mean_token_accuracy": 0.1690695196390152,
      "num_tokens": 30555105.0,
      "step": 14720
    },
    {
      "entropy": 5.777938985824585,
      "epoch": 1.33572206095791,
      "grad_norm": 7.15625,
      "learning_rate": 0.0004826984997959082,
      "loss": 5.5568,
      "mean_token_accuracy": 0.15759443938732148,
      "num_tokens": 30564912.0,
      "step": 14725
    },
    {
      "entropy": 5.77489447593689,
      "epoch": 1.3361756168359942,
      "grad_norm": 2.6875,
      "learning_rate": 0.000482686055008802,
      "loss": 5.5135,
      "mean_token_accuracy": 0.15581004172563553,
      "num_tokens": 30575862.0,
      "step": 14730
    },
    {
      "entropy": 5.834559202194214,
      "epoch": 1.3366291727140784,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048267360592669535,
      "loss": 5.6252,
      "mean_token_accuracy": 0.1580335959792137,
      "num_tokens": 30586951.0,
      "step": 14735
    },
    {
      "entropy": 5.757383251190186,
      "epoch": 1.3370827285921625,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004826611525498456,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16637669056653975,
      "num_tokens": 30597604.0,
      "step": 14740
    },
    {
      "entropy": 5.8102305889129635,
      "epoch": 1.3375362844702468,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004826486948785103,
      "loss": 5.6056,
      "mean_token_accuracy": 0.15691205561161042,
      "num_tokens": 30608157.0,
      "step": 14745
    },
    {
      "entropy": 5.836743450164795,
      "epoch": 1.3379898403483308,
      "grad_norm": 2.375,
      "learning_rate": 0.00048263623291294706,
      "loss": 5.5319,
      "mean_token_accuracy": 0.16227416396141053,
      "num_tokens": 30618527.0,
      "step": 14750
    },
    {
      "entropy": 5.81488676071167,
      "epoch": 1.338443396226415,
      "grad_norm": 3.484375,
      "learning_rate": 0.00048262376665341355,
      "loss": 5.6045,
      "mean_token_accuracy": 0.1574239268898964,
      "num_tokens": 30629253.0,
      "step": 14755
    },
    {
      "entropy": 5.784015655517578,
      "epoch": 1.3388969521044993,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004826112961001677,
      "loss": 5.5685,
      "mean_token_accuracy": 0.14875016510486602,
      "num_tokens": 30639015.0,
      "step": 14760
    },
    {
      "entropy": 5.852496147155762,
      "epoch": 1.3393505079825834,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048259882125346735,
      "loss": 5.643,
      "mean_token_accuracy": 0.15132414400577546,
      "num_tokens": 30650035.0,
      "step": 14765
    },
    {
      "entropy": 5.7967339038848875,
      "epoch": 1.3398040638606676,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048258634211357045,
      "loss": 5.5474,
      "mean_token_accuracy": 0.16385605037212372,
      "num_tokens": 30659813.0,
      "step": 14770
    },
    {
      "entropy": 5.75407223701477,
      "epoch": 1.3402576197387517,
      "grad_norm": 3.0,
      "learning_rate": 0.0004825738586807351,
      "loss": 5.5272,
      "mean_token_accuracy": 0.16010973155498504,
      "num_tokens": 30670243.0,
      "step": 14775
    },
    {
      "entropy": 5.71759181022644,
      "epoch": 1.340711175616836,
      "grad_norm": 4.625,
      "learning_rate": 0.00048256137095521945,
      "loss": 5.4627,
      "mean_token_accuracy": 0.16247102320194245,
      "num_tokens": 30680056.0,
      "step": 14780
    },
    {
      "entropy": 5.802429485321045,
      "epoch": 1.3411647314949202,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048254887893728175,
      "loss": 5.4366,
      "mean_token_accuracy": 0.1645974799990654,
      "num_tokens": 30690536.0,
      "step": 14785
    },
    {
      "entropy": 5.751274347305298,
      "epoch": 1.3416182873730045,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004825363826271804,
      "loss": 5.5115,
      "mean_token_accuracy": 0.15611373335123063,
      "num_tokens": 30700200.0,
      "step": 14790
    },
    {
      "entropy": 5.715616178512573,
      "epoch": 1.3420718432510885,
      "grad_norm": 2.75,
      "learning_rate": 0.0004825238820251737,
      "loss": 5.4132,
      "mean_token_accuracy": 0.16512043476104737,
      "num_tokens": 30710576.0,
      "step": 14795
    },
    {
      "entropy": 5.7390828132629395,
      "epoch": 1.3425253991291728,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048251137713152027,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15795932412147523,
      "num_tokens": 30721495.0,
      "step": 14800
    },
    {
      "entropy": 5.7709251880645756,
      "epoch": 1.3429789550072568,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004824988679464787,
      "loss": 5.5501,
      "mean_token_accuracy": 0.1603291556239128,
      "num_tokens": 30731700.0,
      "step": 14805
    },
    {
      "entropy": 5.8343610763549805,
      "epoch": 1.343432510885341,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048248635447030767,
      "loss": 5.5544,
      "mean_token_accuracy": 0.1580033928155899,
      "num_tokens": 30743421.0,
      "step": 14810
    },
    {
      "entropy": 5.7802070617675785,
      "epoch": 1.3438860667634254,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004824738367032659,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16694968491792678,
      "num_tokens": 30754238.0,
      "step": 14815
    },
    {
      "entropy": 5.792879438400268,
      "epoch": 1.3443396226415094,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004824613146456123,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1638367012143135,
      "num_tokens": 30763938.0,
      "step": 14820
    },
    {
      "entropy": 5.792331266403198,
      "epoch": 1.3447931785195937,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004824487882976059,
      "loss": 5.6102,
      "mean_token_accuracy": 0.15480872020125389,
      "num_tokens": 30773654.0,
      "step": 14825
    },
    {
      "entropy": 5.825184679031372,
      "epoch": 1.3452467343976777,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048243625765950564,
      "loss": 5.5761,
      "mean_token_accuracy": 0.15686182379722596,
      "num_tokens": 30784659.0,
      "step": 14830
    },
    {
      "entropy": 5.736748170852661,
      "epoch": 1.345700290275762,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004824237227315708,
      "loss": 5.4348,
      "mean_token_accuracy": 0.15957257449626921,
      "num_tokens": 30795194.0,
      "step": 14835
    },
    {
      "entropy": 5.62821593284607,
      "epoch": 1.3461538461538463,
      "grad_norm": 3.75,
      "learning_rate": 0.0004824111835140604,
      "loss": 5.3658,
      "mean_token_accuracy": 0.17263796627521516,
      "num_tokens": 30805623.0,
      "step": 14840
    },
    {
      "entropy": 5.704867887496948,
      "epoch": 1.3466074020319303,
      "grad_norm": 2.625,
      "learning_rate": 0.000482398640007234,
      "loss": 5.4387,
      "mean_token_accuracy": 0.16823608204722404,
      "num_tokens": 30814899.0,
      "step": 14845
    },
    {
      "entropy": 5.723225498199463,
      "epoch": 1.3470609579100146,
      "grad_norm": 3.421875,
      "learning_rate": 0.00048238609221135076,
      "loss": 5.5576,
      "mean_token_accuracy": 0.1598539412021637,
      "num_tokens": 30825546.0,
      "step": 14850
    },
    {
      "entropy": 5.689556980133057,
      "epoch": 1.3475145137880986,
      "grad_norm": 3.625,
      "learning_rate": 0.0004823735401266704,
      "loss": 5.441,
      "mean_token_accuracy": 0.1634058475494385,
      "num_tokens": 30836406.0,
      "step": 14855
    },
    {
      "entropy": 5.714275360107422,
      "epoch": 1.3479680696661829,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048236098375345227,
      "loss": 5.5454,
      "mean_token_accuracy": 0.16119794845581054,
      "num_tokens": 30847966.0,
      "step": 14860
    },
    {
      "entropy": 5.814116144180298,
      "epoch": 1.3484216255442671,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004823484230919563,
      "loss": 5.544,
      "mean_token_accuracy": 0.16130483895540237,
      "num_tokens": 30856831.0,
      "step": 14865
    },
    {
      "entropy": 5.684151077270508,
      "epoch": 1.3488751814223512,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048233585814244195,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17132841795682907,
      "num_tokens": 30866076.0,
      "step": 14870
    },
    {
      "entropy": 5.708601236343384,
      "epoch": 1.3493287373004355,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004823232889051693,
      "loss": 5.4424,
      "mean_token_accuracy": 0.1712167575955391,
      "num_tokens": 30876845.0,
      "step": 14875
    },
    {
      "entropy": 5.696187973022461,
      "epoch": 1.3497822931785195,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048231071538039816,
      "loss": 5.5112,
      "mean_token_accuracy": 0.16192753463983536,
      "num_tokens": 30887448.0,
      "step": 14880
    },
    {
      "entropy": 5.791379356384278,
      "epoch": 1.3502358490566038,
      "grad_norm": 6.5625,
      "learning_rate": 0.00048229813756838863,
      "loss": 5.4952,
      "mean_token_accuracy": 0.16222575306892395,
      "num_tokens": 30898414.0,
      "step": 14885
    },
    {
      "entropy": 5.767092227935791,
      "epoch": 1.350689404934688,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004822855554694007,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16521988809108734,
      "num_tokens": 30908559.0,
      "step": 14890
    },
    {
      "entropy": 5.7634485244750975,
      "epoch": 1.351142960812772,
      "grad_norm": 3.953125,
      "learning_rate": 0.00048227296908369475,
      "loss": 5.4981,
      "mean_token_accuracy": 0.16848183274269105,
      "num_tokens": 30918655.0,
      "step": 14895
    },
    {
      "entropy": 5.742977952957153,
      "epoch": 1.3515965166908563,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048226037841153104,
      "loss": 5.5153,
      "mean_token_accuracy": 0.16340841054916383,
      "num_tokens": 30929188.0,
      "step": 14900
    },
    {
      "entropy": 5.780737352371216,
      "epoch": 1.3520500725689404,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004822477834531698,
      "loss": 5.5658,
      "mean_token_accuracy": 0.1629060611128807,
      "num_tokens": 30939902.0,
      "step": 14905
    },
    {
      "entropy": 5.775293445587158,
      "epoch": 1.3525036284470247,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004822351842088716,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16254482120275499,
      "num_tokens": 30951289.0,
      "step": 14910
    },
    {
      "entropy": 5.855059623718262,
      "epoch": 1.352957184325109,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004822225806788969,
      "loss": 5.6355,
      "mean_token_accuracy": 0.15570251047611236,
      "num_tokens": 30961233.0,
      "step": 14915
    },
    {
      "entropy": 5.756930255889893,
      "epoch": 1.353410740203193,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048220997286350654,
      "loss": 5.4639,
      "mean_token_accuracy": 0.16847881823778152,
      "num_tokens": 30971922.0,
      "step": 14920
    },
    {
      "entropy": 5.718548154830932,
      "epoch": 1.3538642960812772,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048219736076296114,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16648878604173661,
      "num_tokens": 30981358.0,
      "step": 14925
    },
    {
      "entropy": 5.762027549743652,
      "epoch": 1.3543178519593613,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004821847443775214,
      "loss": 5.6136,
      "mean_token_accuracy": 0.16171139925718309,
      "num_tokens": 30991984.0,
      "step": 14930
    },
    {
      "entropy": 5.760503959655762,
      "epoch": 1.3547714078374455,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048217212370744853,
      "loss": 5.4896,
      "mean_token_accuracy": 0.16370626986026765,
      "num_tokens": 31002790.0,
      "step": 14935
    },
    {
      "entropy": 5.722923517227173,
      "epoch": 1.3552249637155298,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048215949875300327,
      "loss": 5.3618,
      "mean_token_accuracy": 0.16972851753234863,
      "num_tokens": 31012827.0,
      "step": 14940
    },
    {
      "entropy": 5.709669351577759,
      "epoch": 1.3556785195936139,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004821468695144467,
      "loss": 5.4738,
      "mean_token_accuracy": 0.1623466730117798,
      "num_tokens": 31023075.0,
      "step": 14945
    },
    {
      "entropy": 5.771741914749145,
      "epoch": 1.3561320754716981,
      "grad_norm": 3.71875,
      "learning_rate": 0.00048213423599204015,
      "loss": 5.5134,
      "mean_token_accuracy": 0.16402802467346192,
      "num_tokens": 31032454.0,
      "step": 14950
    },
    {
      "entropy": 5.811146640777588,
      "epoch": 1.3565856313497822,
      "grad_norm": 3.546875,
      "learning_rate": 0.00048212159818604477,
      "loss": 5.6107,
      "mean_token_accuracy": 0.15468652546405792,
      "num_tokens": 31042498.0,
      "step": 14955
    },
    {
      "entropy": 5.725186109542847,
      "epoch": 1.3570391872278664,
      "grad_norm": 2.5625,
      "learning_rate": 0.000482108956096722,
      "loss": 5.5006,
      "mean_token_accuracy": 0.15862648338079452,
      "num_tokens": 31053188.0,
      "step": 14960
    },
    {
      "entropy": 5.7335608959197994,
      "epoch": 1.3574927431059507,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048209630972433326,
      "loss": 5.4812,
      "mean_token_accuracy": 0.17178256064653397,
      "num_tokens": 31062840.0,
      "step": 14965
    },
    {
      "entropy": 5.873023271560669,
      "epoch": 1.357946298984035,
      "grad_norm": 3.59375,
      "learning_rate": 0.00048208365906914,
      "loss": 5.6172,
      "mean_token_accuracy": 0.16384676843881607,
      "num_tokens": 31072646.0,
      "step": 14970
    },
    {
      "entropy": 5.773893594741821,
      "epoch": 1.358399854862119,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048207100413140397,
      "loss": 5.452,
      "mean_token_accuracy": 0.15889616310596466,
      "num_tokens": 31082700.0,
      "step": 14975
    },
    {
      "entropy": 5.780833530426025,
      "epoch": 1.3588534107402033,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004820583449113868,
      "loss": 5.5517,
      "mean_token_accuracy": 0.16151975616812705,
      "num_tokens": 31093743.0,
      "step": 14980
    },
    {
      "entropy": 5.776302528381348,
      "epoch": 1.3593069666182873,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048204568140935023,
      "loss": 5.473,
      "mean_token_accuracy": 0.1655988872051239,
      "num_tokens": 31104293.0,
      "step": 14985
    },
    {
      "entropy": 5.70539174079895,
      "epoch": 1.3597605224963716,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004820330136255563,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1667391374707222,
      "num_tokens": 31114244.0,
      "step": 14990
    },
    {
      "entropy": 5.760334777832031,
      "epoch": 1.3602140783744558,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048202034156026683,
      "loss": 5.4616,
      "mean_token_accuracy": 0.16275528371334075,
      "num_tokens": 31123596.0,
      "step": 14995
    },
    {
      "entropy": 5.763767051696777,
      "epoch": 1.36066763425254,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048200766521374396,
      "loss": 5.5456,
      "mean_token_accuracy": 0.15420500934123993,
      "num_tokens": 31134258.0,
      "step": 15000
    },
    {
      "epoch": 1.36066763425254,
      "eval_entropy": 5.5396002030480505,
      "eval_loss": 5.5773420333862305,
      "eval_mean_token_accuracy": 0.16793524764090506,
      "eval_num_tokens": 31134258.0,
      "eval_runtime": 26.7826,
      "eval_samples_per_second": 1320.26,
      "eval_steps_per_second": 165.033,
      "step": 15000
    },
    {
      "entropy": 5.717752122879029,
      "epoch": 1.3611211901306242,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004819949845862498,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1645461365580559,
      "num_tokens": 31144806.0,
      "step": 15005
    },
    {
      "entropy": 5.8126702308654785,
      "epoch": 1.3615747460087082,
      "grad_norm": 3.0,
      "learning_rate": 0.0004819822996780466,
      "loss": 5.6169,
      "mean_token_accuracy": 0.153952457010746,
      "num_tokens": 31155642.0,
      "step": 15010
    },
    {
      "entropy": 5.822125768661499,
      "epoch": 1.3620283018867925,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004819696104893967,
      "loss": 5.446,
      "mean_token_accuracy": 0.16530997008085252,
      "num_tokens": 31165034.0,
      "step": 15015
    },
    {
      "entropy": 5.759728717803955,
      "epoch": 1.3624818577648767,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004819569170205625,
      "loss": 5.5258,
      "mean_token_accuracy": 0.159354268014431,
      "num_tokens": 31176491.0,
      "step": 15020
    },
    {
      "entropy": 5.78947343826294,
      "epoch": 1.3629354136429608,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048194421927180654,
      "loss": 5.5549,
      "mean_token_accuracy": 0.15823565870523454,
      "num_tokens": 31186198.0,
      "step": 15025
    },
    {
      "entropy": 5.81511869430542,
      "epoch": 1.363388969521045,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004819315172433914,
      "loss": 5.5172,
      "mean_token_accuracy": 0.16630595177412033,
      "num_tokens": 31197094.0,
      "step": 15030
    },
    {
      "entropy": 5.74181547164917,
      "epoch": 1.363842525399129,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048191881093557977,
      "loss": 5.5259,
      "mean_token_accuracy": 0.15497529059648513,
      "num_tokens": 31207505.0,
      "step": 15035
    },
    {
      "entropy": 5.8138697147369385,
      "epoch": 1.3642960812772134,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048190610034863434,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1516956239938736,
      "num_tokens": 31218809.0,
      "step": 15040
    },
    {
      "entropy": 5.820884418487549,
      "epoch": 1.3647496371552976,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004818933854828181,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1510918378829956,
      "num_tokens": 31230281.0,
      "step": 15045
    },
    {
      "entropy": 5.707786512374878,
      "epoch": 1.3652031930333817,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004818806663383939,
      "loss": 5.4464,
      "mean_token_accuracy": 0.17114241868257524,
      "num_tokens": 31239573.0,
      "step": 15050
    },
    {
      "entropy": 5.702888154983521,
      "epoch": 1.365656748911466,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004818679429156248,
      "loss": 5.474,
      "mean_token_accuracy": 0.16589350998401642,
      "num_tokens": 31249671.0,
      "step": 15055
    },
    {
      "entropy": 5.701572322845459,
      "epoch": 1.36611030478955,
      "grad_norm": 2.5,
      "learning_rate": 0.00048185521521477385,
      "loss": 5.5002,
      "mean_token_accuracy": 0.1607659250497818,
      "num_tokens": 31259601.0,
      "step": 15060
    },
    {
      "entropy": 5.763548374176025,
      "epoch": 1.3665638606676342,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004818424832361044,
      "loss": 5.4824,
      "mean_token_accuracy": 0.16547009497880935,
      "num_tokens": 31269085.0,
      "step": 15065
    },
    {
      "entropy": 5.719637870788574,
      "epoch": 1.3670174165457185,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004818297469798797,
      "loss": 5.5037,
      "mean_token_accuracy": 0.15762878209352493,
      "num_tokens": 31280177.0,
      "step": 15070
    },
    {
      "entropy": 5.76025357246399,
      "epoch": 1.3674709724238026,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048181700644636314,
      "loss": 5.4286,
      "mean_token_accuracy": 0.1678305059671402,
      "num_tokens": 31290545.0,
      "step": 15075
    },
    {
      "entropy": 5.766378498077392,
      "epoch": 1.3679245283018868,
      "grad_norm": 2.75,
      "learning_rate": 0.00048180426163581816,
      "loss": 5.4081,
      "mean_token_accuracy": 0.16334231793880463,
      "num_tokens": 31301143.0,
      "step": 15080
    },
    {
      "entropy": 5.723943281173706,
      "epoch": 1.3683780841799709,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004817915125485083,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16338462829589845,
      "num_tokens": 31311071.0,
      "step": 15085
    },
    {
      "entropy": 5.781814670562744,
      "epoch": 1.3688316400580551,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048177875918469735,
      "loss": 5.5754,
      "mean_token_accuracy": 0.1601577639579773,
      "num_tokens": 31322263.0,
      "step": 15090
    },
    {
      "entropy": 5.762464141845703,
      "epoch": 1.3692851959361394,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048176600154464897,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16564786732196807,
      "num_tokens": 31332975.0,
      "step": 15095
    },
    {
      "entropy": 5.852451992034912,
      "epoch": 1.3697387518142234,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004817532396286269,
      "loss": 5.6146,
      "mean_token_accuracy": 0.15478328615427017,
      "num_tokens": 31345252.0,
      "step": 15100
    },
    {
      "entropy": 5.761891651153564,
      "epoch": 1.3701923076923077,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004817404734368952,
      "loss": 5.4541,
      "mean_token_accuracy": 0.16651802211999894,
      "num_tokens": 31355479.0,
      "step": 15105
    },
    {
      "entropy": 5.803439044952393,
      "epoch": 1.3706458635703918,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048172770296971785,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1570475608110428,
      "num_tokens": 31366020.0,
      "step": 15110
    },
    {
      "entropy": 5.801712989807129,
      "epoch": 1.371099419448476,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004817149282273589,
      "loss": 5.503,
      "mean_token_accuracy": 0.16579040586948396,
      "num_tokens": 31376055.0,
      "step": 15115
    },
    {
      "entropy": 5.767214727401734,
      "epoch": 1.3715529753265603,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004817021492100825,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1560145229101181,
      "num_tokens": 31386179.0,
      "step": 15120
    },
    {
      "entropy": 5.780704689025879,
      "epoch": 1.3720065312046443,
      "grad_norm": 2.796875,
      "learning_rate": 0.000481689365918153,
      "loss": 5.4428,
      "mean_token_accuracy": 0.16725848615169525,
      "num_tokens": 31395850.0,
      "step": 15125
    },
    {
      "entropy": 5.7755176544189455,
      "epoch": 1.3724600870827286,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004816765783518348,
      "loss": 5.4158,
      "mean_token_accuracy": 0.16963386982679368,
      "num_tokens": 31405053.0,
      "step": 15130
    },
    {
      "entropy": 5.7232893943786625,
      "epoch": 1.3729136429608126,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004816637865113923,
      "loss": 5.4347,
      "mean_token_accuracy": 0.17319557517766954,
      "num_tokens": 31415153.0,
      "step": 15135
    },
    {
      "entropy": 5.717486095428467,
      "epoch": 1.373367198838897,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048165099039708994,
      "loss": 5.5361,
      "mean_token_accuracy": 0.16394708454608917,
      "num_tokens": 31425451.0,
      "step": 15140
    },
    {
      "entropy": 5.705425643920899,
      "epoch": 1.3738207547169812,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004816381900091925,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16095294654369355,
      "num_tokens": 31436298.0,
      "step": 15145
    },
    {
      "entropy": 5.7616286277771,
      "epoch": 1.3742743105950654,
      "grad_norm": 3.515625,
      "learning_rate": 0.00048162538534796465,
      "loss": 5.4954,
      "mean_token_accuracy": 0.1672042801976204,
      "num_tokens": 31446932.0,
      "step": 15150
    },
    {
      "entropy": 5.802530288696289,
      "epoch": 1.3747278664731495,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004816125764136711,
      "loss": 5.5663,
      "mean_token_accuracy": 0.14875692799687384,
      "num_tokens": 31457582.0,
      "step": 15155
    },
    {
      "entropy": 5.835864400863647,
      "epoch": 1.3751814223512338,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004815997632065769,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15566431134939193,
      "num_tokens": 31468168.0,
      "step": 15160
    },
    {
      "entropy": 5.767144775390625,
      "epoch": 1.3756349782293178,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048158694572694694,
      "loss": 5.5181,
      "mean_token_accuracy": 0.1651332423090935,
      "num_tokens": 31478524.0,
      "step": 15165
    },
    {
      "entropy": 5.6943634986877445,
      "epoch": 1.376088534107402,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048157412397504623,
      "loss": 5.4963,
      "mean_token_accuracy": 0.16523564755916595,
      "num_tokens": 31488608.0,
      "step": 15170
    },
    {
      "entropy": 5.718742370605469,
      "epoch": 1.3765420899854863,
      "grad_norm": 2.125,
      "learning_rate": 0.0004815612979511401,
      "loss": 5.5155,
      "mean_token_accuracy": 0.16770175993442535,
      "num_tokens": 31498984.0,
      "step": 15175
    },
    {
      "entropy": 5.8681886196136475,
      "epoch": 1.3769956458635704,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048154846765549353,
      "loss": 5.5393,
      "mean_token_accuracy": 0.16252620071172713,
      "num_tokens": 31509110.0,
      "step": 15180
    },
    {
      "entropy": 5.724084186553955,
      "epoch": 1.3774492017416546,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048153563308837215,
      "loss": 5.511,
      "mean_token_accuracy": 0.1623568817973137,
      "num_tokens": 31519266.0,
      "step": 15185
    },
    {
      "entropy": 5.740420818328857,
      "epoch": 1.3779027576197387,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048152279425004124,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16346994936466216,
      "num_tokens": 31529043.0,
      "step": 15190
    },
    {
      "entropy": 5.813358926773072,
      "epoch": 1.378356313497823,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004815099511407663,
      "loss": 5.5121,
      "mean_token_accuracy": 0.16478234231472016,
      "num_tokens": 31538650.0,
      "step": 15195
    },
    {
      "entropy": 5.631648349761963,
      "epoch": 1.3788098693759072,
      "grad_norm": 1.90625,
      "learning_rate": 0.000481497103760813,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16664578318595885,
      "num_tokens": 31549392.0,
      "step": 15200
    },
    {
      "entropy": 5.761192512512207,
      "epoch": 1.3792634252539913,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048148425211044684,
      "loss": 5.6367,
      "mean_token_accuracy": 0.15400453358888627,
      "num_tokens": 31559999.0,
      "step": 15205
    },
    {
      "entropy": 5.897281455993652,
      "epoch": 1.3797169811320755,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004814713961899338,
      "loss": 5.5233,
      "mean_token_accuracy": 0.15898694843053818,
      "num_tokens": 31570108.0,
      "step": 15210
    },
    {
      "entropy": 5.734611463546753,
      "epoch": 1.3801705370101596,
      "grad_norm": 3.796875,
      "learning_rate": 0.00048145853599953966,
      "loss": 5.4648,
      "mean_token_accuracy": 0.16116732805967332,
      "num_tokens": 31579698.0,
      "step": 15215
    },
    {
      "entropy": 5.641057300567627,
      "epoch": 1.3806240928882438,
      "grad_norm": 3.6875,
      "learning_rate": 0.00048144567153953036,
      "loss": 5.533,
      "mean_token_accuracy": 0.1598899021744728,
      "num_tokens": 31589292.0,
      "step": 15220
    },
    {
      "entropy": 5.782537269592285,
      "epoch": 1.381077648766328,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048143280281017197,
      "loss": 5.5834,
      "mean_token_accuracy": 0.16192115247249603,
      "num_tokens": 31599848.0,
      "step": 15225
    },
    {
      "entropy": 5.801352071762085,
      "epoch": 1.3815312046444121,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004814199298117307,
      "loss": 5.5446,
      "mean_token_accuracy": 0.16301317363977433,
      "num_tokens": 31611109.0,
      "step": 15230
    },
    {
      "entropy": 5.767622280120849,
      "epoch": 1.3819847605224964,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048140705254447256,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16412929445505142,
      "num_tokens": 31621140.0,
      "step": 15235
    },
    {
      "entropy": 5.8224671363830565,
      "epoch": 1.3824383164005805,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048139417100866405,
      "loss": 5.5618,
      "mean_token_accuracy": 0.16204821020364762,
      "num_tokens": 31631268.0,
      "step": 15240
    },
    {
      "entropy": 5.785569381713867,
      "epoch": 1.3828918722786647,
      "grad_norm": 4.0,
      "learning_rate": 0.00048138128520457146,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16300802081823348,
      "num_tokens": 31640758.0,
      "step": 15245
    },
    {
      "entropy": 5.65971155166626,
      "epoch": 1.383345428156749,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004813683951324613,
      "loss": 5.4973,
      "mean_token_accuracy": 0.15803689658641815,
      "num_tokens": 31651060.0,
      "step": 15250
    },
    {
      "entropy": 5.811180877685547,
      "epoch": 1.383798984034833,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048135550079260014,
      "loss": 5.6075,
      "mean_token_accuracy": 0.1579552635550499,
      "num_tokens": 31662282.0,
      "step": 15255
    },
    {
      "entropy": 5.80939416885376,
      "epoch": 1.3842525399129173,
      "grad_norm": 2.5,
      "learning_rate": 0.0004813426021852547,
      "loss": 5.474,
      "mean_token_accuracy": 0.16040838062763213,
      "num_tokens": 31672364.0,
      "step": 15260
    },
    {
      "entropy": 5.7709283351898195,
      "epoch": 1.3847060957910013,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048132969931069156,
      "loss": 5.5424,
      "mean_token_accuracy": 0.15810130089521407,
      "num_tokens": 31683094.0,
      "step": 15265
    },
    {
      "entropy": 5.721621370315551,
      "epoch": 1.3851596516690856,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004813167921691777,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16374632716178894,
      "num_tokens": 31693087.0,
      "step": 15270
    },
    {
      "entropy": 5.772472476959228,
      "epoch": 1.3856132075471699,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048130388076098003,
      "loss": 5.4976,
      "mean_token_accuracy": 0.16079585552215575,
      "num_tokens": 31703708.0,
      "step": 15275
    },
    {
      "entropy": 5.735334396362305,
      "epoch": 1.386066763425254,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048129096508636547,
      "loss": 5.4262,
      "mean_token_accuracy": 0.16598720848560333,
      "num_tokens": 31713640.0,
      "step": 15280
    },
    {
      "entropy": 5.715126657485962,
      "epoch": 1.3865203193033382,
      "grad_norm": 2.375,
      "learning_rate": 0.00048127804514560126,
      "loss": 5.5138,
      "mean_token_accuracy": 0.16181522458791733,
      "num_tokens": 31723526.0,
      "step": 15285
    },
    {
      "entropy": 5.700529479980469,
      "epoch": 1.3869738751814222,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048126512093895447,
      "loss": 5.4617,
      "mean_token_accuracy": 0.16449553668498992,
      "num_tokens": 31733265.0,
      "step": 15290
    },
    {
      "entropy": 5.806404876708984,
      "epoch": 1.3874274310595065,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004812521924666925,
      "loss": 5.4675,
      "mean_token_accuracy": 0.16545580625534057,
      "num_tokens": 31743832.0,
      "step": 15295
    },
    {
      "entropy": 5.776945161819458,
      "epoch": 1.3878809869375908,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004812392597290825,
      "loss": 5.5113,
      "mean_token_accuracy": 0.17137772291898729,
      "num_tokens": 31754443.0,
      "step": 15300
    },
    {
      "entropy": 5.703382873535157,
      "epoch": 1.3883345428156748,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004812263227263921,
      "loss": 5.4802,
      "mean_token_accuracy": 0.1649916037917137,
      "num_tokens": 31765304.0,
      "step": 15305
    },
    {
      "entropy": 5.730464696884155,
      "epoch": 1.388788098693759,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004812133814588888,
      "loss": 5.5066,
      "mean_token_accuracy": 0.15905141830444336,
      "num_tokens": 31775809.0,
      "step": 15310
    },
    {
      "entropy": 5.774436569213867,
      "epoch": 1.3892416545718431,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004812004359268402,
      "loss": 5.4924,
      "mean_token_accuracy": 0.16854310482740403,
      "num_tokens": 31786271.0,
      "step": 15315
    },
    {
      "entropy": 5.752078485488892,
      "epoch": 1.3896952104499274,
      "grad_norm": 3.0,
      "learning_rate": 0.0004811874861305141,
      "loss": 5.5463,
      "mean_token_accuracy": 0.16166966706514357,
      "num_tokens": 31797358.0,
      "step": 15320
    },
    {
      "entropy": 5.729553747177124,
      "epoch": 1.3901487663280117,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004811745320701782,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16523596793413162,
      "num_tokens": 31808303.0,
      "step": 15325
    },
    {
      "entropy": 5.730534648895263,
      "epoch": 1.390602322206096,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004811615737461004,
      "loss": 5.487,
      "mean_token_accuracy": 0.16430894285440445,
      "num_tokens": 31818837.0,
      "step": 15330
    },
    {
      "entropy": 5.818807172775268,
      "epoch": 1.39105587808418,
      "grad_norm": 2.5,
      "learning_rate": 0.00048114861115854877,
      "loss": 5.4956,
      "mean_token_accuracy": 0.16531778275966644,
      "num_tokens": 31829729.0,
      "step": 15335
    },
    {
      "entropy": 5.686241674423218,
      "epoch": 1.3915094339622642,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048113564430779136,
      "loss": 5.4269,
      "mean_token_accuracy": 0.15887285470962526,
      "num_tokens": 31839996.0,
      "step": 15340
    },
    {
      "entropy": 5.665463924407959,
      "epoch": 1.3919629898403483,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004811226731940962,
      "loss": 5.4315,
      "mean_token_accuracy": 0.16976548731327057,
      "num_tokens": 31850298.0,
      "step": 15345
    },
    {
      "entropy": 5.797342920303345,
      "epoch": 1.3924165457184325,
      "grad_norm": 2.75,
      "learning_rate": 0.00048110969781773173,
      "loss": 5.5258,
      "mean_token_accuracy": 0.1703367128968239,
      "num_tokens": 31862170.0,
      "step": 15350
    },
    {
      "entropy": 5.810710859298706,
      "epoch": 1.3928701015965168,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048109671817896614,
      "loss": 5.5592,
      "mean_token_accuracy": 0.15983250439167024,
      "num_tokens": 31872081.0,
      "step": 15355
    },
    {
      "entropy": 5.701780700683594,
      "epoch": 1.3933236574746009,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048108373427806797,
      "loss": 5.5085,
      "mean_token_accuracy": 0.16644221693277358,
      "num_tokens": 31882715.0,
      "step": 15360
    },
    {
      "entropy": 5.73279070854187,
      "epoch": 1.3937772133526851,
      "grad_norm": 2.375,
      "learning_rate": 0.00048107074611530555,
      "loss": 5.4696,
      "mean_token_accuracy": 0.17084608525037764,
      "num_tokens": 31891925.0,
      "step": 15365
    },
    {
      "entropy": 5.805791759490967,
      "epoch": 1.3942307692307692,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048105775369094756,
      "loss": 5.5172,
      "mean_token_accuracy": 0.1582521915435791,
      "num_tokens": 31902320.0,
      "step": 15370
    },
    {
      "entropy": 5.739030838012695,
      "epoch": 1.3946843251088534,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004810447570052628,
      "loss": 5.4184,
      "mean_token_accuracy": 0.1674148589372635,
      "num_tokens": 31912726.0,
      "step": 15375
    },
    {
      "entropy": 5.758109903335571,
      "epoch": 1.3951378809869377,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048103175605851997,
      "loss": 5.5186,
      "mean_token_accuracy": 0.16232790350914,
      "num_tokens": 31924039.0,
      "step": 15380
    },
    {
      "entropy": 5.631223726272583,
      "epoch": 1.3955914368650217,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004810187508509879,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17693812847137452,
      "num_tokens": 31933280.0,
      "step": 15385
    },
    {
      "entropy": 5.7287993907928465,
      "epoch": 1.396044992743106,
      "grad_norm": 6.09375,
      "learning_rate": 0.0004810057413829355,
      "loss": 5.5963,
      "mean_token_accuracy": 0.1629786342382431,
      "num_tokens": 31943182.0,
      "step": 15390
    },
    {
      "entropy": 5.846837711334229,
      "epoch": 1.39649854862119,
      "grad_norm": 3.96875,
      "learning_rate": 0.00048099272765463193,
      "loss": 5.5286,
      "mean_token_accuracy": 0.16220448911190033,
      "num_tokens": 31953256.0,
      "step": 15395
    },
    {
      "entropy": 5.874432802200317,
      "epoch": 1.3969521044992743,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004809797096663463,
      "loss": 5.6291,
      "mean_token_accuracy": 0.15732863992452623,
      "num_tokens": 31964231.0,
      "step": 15400
    },
    {
      "entropy": 5.767672204971314,
      "epoch": 1.3974056603773586,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004809666874183477,
      "loss": 5.4526,
      "mean_token_accuracy": 0.16081662178039552,
      "num_tokens": 31974322.0,
      "step": 15405
    },
    {
      "entropy": 5.773267316818237,
      "epoch": 1.3978592162554426,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004809536609109056,
      "loss": 5.5397,
      "mean_token_accuracy": 0.16579921096563338,
      "num_tokens": 31984239.0,
      "step": 15410
    },
    {
      "entropy": 5.719849538803101,
      "epoch": 1.398312772133527,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048094063014428925,
      "loss": 5.5103,
      "mean_token_accuracy": 0.159497931599617,
      "num_tokens": 31994366.0,
      "step": 15415
    },
    {
      "entropy": 5.814095067977905,
      "epoch": 1.398766328011611,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004809275951187682,
      "loss": 5.5777,
      "mean_token_accuracy": 0.15905752182006835,
      "num_tokens": 32004921.0,
      "step": 15420
    },
    {
      "entropy": 5.757103395462036,
      "epoch": 1.3992198838896952,
      "grad_norm": 3.671875,
      "learning_rate": 0.00048091455583461204,
      "loss": 5.5283,
      "mean_token_accuracy": 0.16185469776391984,
      "num_tokens": 32016385.0,
      "step": 15425
    },
    {
      "entropy": 5.774538183212281,
      "epoch": 1.3996734397677795,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048090151229209036,
      "loss": 5.5224,
      "mean_token_accuracy": 0.15849438756704332,
      "num_tokens": 32027398.0,
      "step": 15430
    },
    {
      "entropy": 5.757079696655273,
      "epoch": 1.4001269956458635,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004808884644914729,
      "loss": 5.5093,
      "mean_token_accuracy": 0.1624175414443016,
      "num_tokens": 32038326.0,
      "step": 15435
    },
    {
      "entropy": 5.722050046920776,
      "epoch": 1.4005805515239478,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004808754124330296,
      "loss": 5.5658,
      "mean_token_accuracy": 0.1636267453432083,
      "num_tokens": 32048744.0,
      "step": 15440
    },
    {
      "entropy": 5.728003597259521,
      "epoch": 1.4010341074020318,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004808623561170302,
      "loss": 5.456,
      "mean_token_accuracy": 0.16920295655727385,
      "num_tokens": 32059133.0,
      "step": 15445
    },
    {
      "entropy": 5.790263509750366,
      "epoch": 1.401487663280116,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004808492955437449,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16421533972024918,
      "num_tokens": 32069277.0,
      "step": 15450
    },
    {
      "entropy": 5.812860822677612,
      "epoch": 1.4019412191582004,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048083623071344374,
      "loss": 5.4377,
      "mean_token_accuracy": 0.16508173942565918,
      "num_tokens": 32079802.0,
      "step": 15455
    },
    {
      "entropy": 5.762965297698974,
      "epoch": 1.4023947750362844,
      "grad_norm": 6.65625,
      "learning_rate": 0.00048082316162639685,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1591537043452263,
      "num_tokens": 32090962.0,
      "step": 15460
    },
    {
      "entropy": 5.664652299880982,
      "epoch": 1.4028483309143687,
      "grad_norm": 3.375,
      "learning_rate": 0.00048081008828287447,
      "loss": 5.5049,
      "mean_token_accuracy": 0.16605946272611619,
      "num_tokens": 32101058.0,
      "step": 15465
    },
    {
      "entropy": 5.820835256576538,
      "epoch": 1.4033018867924527,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048079701068314706,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1669479191303253,
      "num_tokens": 32111763.0,
      "step": 15470
    },
    {
      "entropy": 5.843649196624756,
      "epoch": 1.403755442670537,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048078392882748507,
      "loss": 5.5959,
      "mean_token_accuracy": 0.16264702379703522,
      "num_tokens": 32123556.0,
      "step": 15475
    },
    {
      "entropy": 5.740909719467163,
      "epoch": 1.4042089985486212,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048077084271615895,
      "loss": 5.4767,
      "mean_token_accuracy": 0.16514982879161835,
      "num_tokens": 32134012.0,
      "step": 15480
    },
    {
      "entropy": 5.6725279808044435,
      "epoch": 1.4046625544267053,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004807577523494394,
      "loss": 5.4522,
      "mean_token_accuracy": 0.17187827229499816,
      "num_tokens": 32144540.0,
      "step": 15485
    },
    {
      "entropy": 5.769112539291382,
      "epoch": 1.4051161103047896,
      "grad_norm": 2.25,
      "learning_rate": 0.00048074465772759707,
      "loss": 5.5574,
      "mean_token_accuracy": 0.16384157985448838,
      "num_tokens": 32154655.0,
      "step": 15490
    },
    {
      "entropy": 5.796892356872559,
      "epoch": 1.4055696661828736,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048073155885090276,
      "loss": 5.4748,
      "mean_token_accuracy": 0.16377089619636537,
      "num_tokens": 32165191.0,
      "step": 15495
    },
    {
      "entropy": 5.797755289077759,
      "epoch": 1.4060232220609579,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048071845571962733,
      "loss": 5.5422,
      "mean_token_accuracy": 0.15379638820886612,
      "num_tokens": 32175961.0,
      "step": 15500
    },
    {
      "entropy": 5.794145584106445,
      "epoch": 1.4064767779390421,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004807053483340419,
      "loss": 5.5014,
      "mean_token_accuracy": 0.16253344863653182,
      "num_tokens": 32187073.0,
      "step": 15505
    },
    {
      "entropy": 5.766656684875488,
      "epoch": 1.4069303338171264,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004806922366944175,
      "loss": 5.5177,
      "mean_token_accuracy": 0.1663169279694557,
      "num_tokens": 32197426.0,
      "step": 15510
    },
    {
      "entropy": 5.803904247283936,
      "epoch": 1.4073838896952104,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048067912080102513,
      "loss": 5.4842,
      "mean_token_accuracy": 0.17292481213808059,
      "num_tokens": 32206847.0,
      "step": 15515
    },
    {
      "entropy": 5.669209146499634,
      "epoch": 1.4078374455732947,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004806660006541362,
      "loss": 5.5287,
      "mean_token_accuracy": 0.16585462987422944,
      "num_tokens": 32217647.0,
      "step": 15520
    },
    {
      "entropy": 5.691448640823364,
      "epoch": 1.4082910014513788,
      "grad_norm": 12.125,
      "learning_rate": 0.0004806528762540219,
      "loss": 5.3849,
      "mean_token_accuracy": 0.17399238049983978,
      "num_tokens": 32227133.0,
      "step": 15525
    },
    {
      "entropy": 5.79808201789856,
      "epoch": 1.408744557329463,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048063974760095374,
      "loss": 5.592,
      "mean_token_accuracy": 0.1597960263490677,
      "num_tokens": 32237378.0,
      "step": 15530
    },
    {
      "entropy": 5.7014007568359375,
      "epoch": 1.4091981132075473,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048062661469520327,
      "loss": 5.4615,
      "mean_token_accuracy": 0.16750411540269852,
      "num_tokens": 32247213.0,
      "step": 15535
    },
    {
      "entropy": 5.809665250778198,
      "epoch": 1.4096516690856313,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048061347753704193,
      "loss": 5.5124,
      "mean_token_accuracy": 0.16361294835805892,
      "num_tokens": 32257100.0,
      "step": 15540
    },
    {
      "entropy": 5.79362211227417,
      "epoch": 1.4101052249637156,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004806003361267415,
      "loss": 5.4763,
      "mean_token_accuracy": 0.15660134255886077,
      "num_tokens": 32267456.0,
      "step": 15545
    },
    {
      "entropy": 5.673174905776977,
      "epoch": 1.4105587808417996,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004805871904645737,
      "loss": 5.455,
      "mean_token_accuracy": 0.16839295178651809,
      "num_tokens": 32278220.0,
      "step": 15550
    },
    {
      "entropy": 5.620787668228149,
      "epoch": 1.411012336719884,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004805740405508103,
      "loss": 5.5114,
      "mean_token_accuracy": 0.16842640936374664,
      "num_tokens": 32289036.0,
      "step": 15555
    },
    {
      "entropy": 5.696955060958862,
      "epoch": 1.4114658925979682,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004805608863857236,
      "loss": 5.4198,
      "mean_token_accuracy": 0.16499495804309844,
      "num_tokens": 32298909.0,
      "step": 15560
    },
    {
      "entropy": 5.768050622940064,
      "epoch": 1.4119194484760522,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.4805,
      "mean_token_accuracy": 0.1650126025080681,
      "num_tokens": 32309851.0,
      "step": 15565
    },
    {
      "entropy": 5.705271244049072,
      "epoch": 1.4123730043541365,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004805345653026675,
      "loss": 5.4106,
      "mean_token_accuracy": 0.17279548943042755,
      "num_tokens": 32319336.0,
      "step": 15570
    },
    {
      "entropy": 5.767312240600586,
      "epoch": 1.4128265602322205,
      "grad_norm": 5.25,
      "learning_rate": 0.00048052139838524263,
      "loss": 5.505,
      "mean_token_accuracy": 0.16423501521348954,
      "num_tokens": 32329898.0,
      "step": 15575
    },
    {
      "entropy": 5.738998317718506,
      "epoch": 1.4132801161103048,
      "grad_norm": 2.125,
      "learning_rate": 0.00048050822721758274,
      "loss": 5.3624,
      "mean_token_accuracy": 0.16864762753248214,
      "num_tokens": 32340100.0,
      "step": 15580
    },
    {
      "entropy": 5.760234260559082,
      "epoch": 1.413733671988389,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004804950517999604,
      "loss": 5.4894,
      "mean_token_accuracy": 0.16327396929264068,
      "num_tokens": 32349663.0,
      "step": 15585
    },
    {
      "entropy": 5.751736497879028,
      "epoch": 1.414187227866473,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004804818721326481,
      "loss": 5.5439,
      "mean_token_accuracy": 0.16417409777641295,
      "num_tokens": 32360461.0,
      "step": 15590
    },
    {
      "entropy": 5.704200077056885,
      "epoch": 1.4146407837445574,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004804686882159183,
      "loss": 5.5224,
      "mean_token_accuracy": 0.16720327287912368,
      "num_tokens": 32370910.0,
      "step": 15595
    },
    {
      "entropy": 5.859882116317749,
      "epoch": 1.4150943396226414,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048045550005004355,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16832513362169266,
      "num_tokens": 32381669.0,
      "step": 15600
    },
    {
      "entropy": 5.7482658386230465,
      "epoch": 1.4155478955007257,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004804423076352968,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16058287173509597,
      "num_tokens": 32392024.0,
      "step": 15605
    },
    {
      "entropy": 5.665823173522949,
      "epoch": 1.41600145137881,
      "grad_norm": 3.125,
      "learning_rate": 0.00048042911097195075,
      "loss": 5.4828,
      "mean_token_accuracy": 0.166927707195282,
      "num_tokens": 32401872.0,
      "step": 15610
    },
    {
      "entropy": 5.723237133026123,
      "epoch": 1.416455007256894,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004804159100602784,
      "loss": 5.434,
      "mean_token_accuracy": 0.1660824939608574,
      "num_tokens": 32411749.0,
      "step": 15615
    },
    {
      "entropy": 5.819490146636963,
      "epoch": 1.4169085631349783,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048040270490055264,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15859984904527663,
      "num_tokens": 32421358.0,
      "step": 15620
    },
    {
      "entropy": 5.81111741065979,
      "epoch": 1.4173621190130623,
      "grad_norm": 2.875,
      "learning_rate": 0.0004803894954930465,
      "loss": 5.517,
      "mean_token_accuracy": 0.16614200472831725,
      "num_tokens": 32432322.0,
      "step": 15625
    },
    {
      "entropy": 5.706794691085816,
      "epoch": 1.4178156748911466,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004803762818380334,
      "loss": 5.5184,
      "mean_token_accuracy": 0.16137531846761705,
      "num_tokens": 32440930.0,
      "step": 15630
    },
    {
      "entropy": 5.769697904586792,
      "epoch": 1.4182692307692308,
      "grad_norm": 4.34375,
      "learning_rate": 0.00048036306393578644,
      "loss": 5.5413,
      "mean_token_accuracy": 0.16021281331777573,
      "num_tokens": 32451124.0,
      "step": 15635
    },
    {
      "entropy": 5.7712822437286375,
      "epoch": 1.4187227866473149,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004803498417865789,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1634540468454361,
      "num_tokens": 32460636.0,
      "step": 15640
    },
    {
      "entropy": 5.77478141784668,
      "epoch": 1.4191763425253991,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004803366153906844,
      "loss": 5.5337,
      "mean_token_accuracy": 0.16628656536340714,
      "num_tokens": 32470501.0,
      "step": 15645
    },
    {
      "entropy": 5.806349277496338,
      "epoch": 1.4196298984034832,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004803233847483763,
      "loss": 5.6444,
      "mean_token_accuracy": 0.15793873965740204,
      "num_tokens": 32481559.0,
      "step": 15650
    },
    {
      "entropy": 5.937090539932251,
      "epoch": 1.4200834542815675,
      "grad_norm": 3.0,
      "learning_rate": 0.00048031014985992825,
      "loss": 5.6896,
      "mean_token_accuracy": 0.15515884682536124,
      "num_tokens": 32493000.0,
      "step": 15655
    },
    {
      "entropy": 5.805704355239868,
      "epoch": 1.4205370101596517,
      "grad_norm": 4.40625,
      "learning_rate": 0.000480296910725614,
      "loss": 5.4682,
      "mean_token_accuracy": 0.17000869363546373,
      "num_tokens": 32502773.0,
      "step": 15660
    },
    {
      "entropy": 5.736756992340088,
      "epoch": 1.4209905660377358,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004802836673457074,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1614047408103943,
      "num_tokens": 32513430.0,
      "step": 15665
    },
    {
      "entropy": 5.703171348571777,
      "epoch": 1.42144412191582,
      "grad_norm": 5.21875,
      "learning_rate": 0.00048027041972048225,
      "loss": 5.4702,
      "mean_token_accuracy": 0.16181192249059678,
      "num_tokens": 32524409.0,
      "step": 15670
    },
    {
      "entropy": 5.737373638153076,
      "epoch": 1.421897677793904,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004802571678502124,
      "loss": 5.5417,
      "mean_token_accuracy": 0.15788987427949905,
      "num_tokens": 32534696.0,
      "step": 15675
    },
    {
      "entropy": 5.820654678344726,
      "epoch": 1.4223512336719883,
      "grad_norm": 3.015625,
      "learning_rate": 0.000480243911735172,
      "loss": 5.5179,
      "mean_token_accuracy": 0.1695757269859314,
      "num_tokens": 32544782.0,
      "step": 15680
    },
    {
      "entropy": 5.780317401885986,
      "epoch": 1.4228047895500726,
      "grad_norm": 2.625,
      "learning_rate": 0.00048023065137563534,
      "loss": 5.5073,
      "mean_token_accuracy": 0.1570200055837631,
      "num_tokens": 32554818.0,
      "step": 15685
    },
    {
      "entropy": 5.769607830047607,
      "epoch": 1.4232583454281569,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004802173867718764,
      "loss": 5.528,
      "mean_token_accuracy": 0.16996509879827498,
      "num_tokens": 32565007.0,
      "step": 15690
    },
    {
      "entropy": 5.77475152015686,
      "epoch": 1.423711901306241,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004802041179241697,
      "loss": 5.495,
      "mean_token_accuracy": 0.16458241194486617,
      "num_tokens": 32575657.0,
      "step": 15695
    },
    {
      "entropy": 5.735434341430664,
      "epoch": 1.4241654571843252,
      "grad_norm": 2.75,
      "learning_rate": 0.0004801908448327895,
      "loss": 5.5352,
      "mean_token_accuracy": 0.16098432689905168,
      "num_tokens": 32586676.0,
      "step": 15700
    },
    {
      "entropy": 5.733594036102295,
      "epoch": 1.4246190130624092,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004801775674980103,
      "loss": 5.483,
      "mean_token_accuracy": 0.16255193799734116,
      "num_tokens": 32596191.0,
      "step": 15705
    },
    {
      "entropy": 5.806463813781738,
      "epoch": 1.4250725689404935,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048016428592010676,
      "loss": 5.5602,
      "mean_token_accuracy": 0.1583818107843399,
      "num_tokens": 32607178.0,
      "step": 15710
    },
    {
      "entropy": 5.824058914184571,
      "epoch": 1.4255261248185778,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048015100009935347,
      "loss": 5.4589,
      "mean_token_accuracy": 0.16543003916740417,
      "num_tokens": 32617301.0,
      "step": 15715
    },
    {
      "entropy": 5.723859119415283,
      "epoch": 1.4259796806966618,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004801377100360253,
      "loss": 5.4605,
      "mean_token_accuracy": 0.16822518557310104,
      "num_tokens": 32627395.0,
      "step": 15720
    },
    {
      "entropy": 5.699683284759521,
      "epoch": 1.426433236574746,
      "grad_norm": 2.5625,
      "learning_rate": 0.000480124415730397,
      "loss": 5.505,
      "mean_token_accuracy": 0.16699720323085784,
      "num_tokens": 32638460.0,
      "step": 15725
    },
    {
      "entropy": 5.7447268009185795,
      "epoch": 1.4268867924528301,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004801111171827434,
      "loss": 5.5716,
      "mean_token_accuracy": 0.16277801394462585,
      "num_tokens": 32649319.0,
      "step": 15730
    },
    {
      "entropy": 5.7271734237670895,
      "epoch": 1.4273403483309144,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004800978143933398,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16398975402116775,
      "num_tokens": 32659976.0,
      "step": 15735
    },
    {
      "entropy": 5.750333786010742,
      "epoch": 1.4277939042089987,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048008450736246094,
      "loss": 5.4106,
      "mean_token_accuracy": 0.16402670443058015,
      "num_tokens": 32670772.0,
      "step": 15740
    },
    {
      "entropy": 5.793039226531983,
      "epoch": 1.4282474600870827,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048007119609038237,
      "loss": 5.5775,
      "mean_token_accuracy": 0.1519896999001503,
      "num_tokens": 32682368.0,
      "step": 15745
    },
    {
      "entropy": 5.692071342468262,
      "epoch": 1.428701015965167,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048005788057737907,
      "loss": 5.4048,
      "mean_token_accuracy": 0.16330707520246507,
      "num_tokens": 32692649.0,
      "step": 15750
    },
    {
      "entropy": 5.797497320175171,
      "epoch": 1.429154571843251,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004800445608237266,
      "loss": 5.6273,
      "mean_token_accuracy": 0.1640828490257263,
      "num_tokens": 32704075.0,
      "step": 15755
    },
    {
      "entropy": 5.798878812789917,
      "epoch": 1.4296081277213353,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048003123682970045,
      "loss": 5.6124,
      "mean_token_accuracy": 0.15504651367664338,
      "num_tokens": 32714670.0,
      "step": 15760
    },
    {
      "entropy": 5.766755867004394,
      "epoch": 1.4300616835994195,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048001790859557596,
      "loss": 5.4534,
      "mean_token_accuracy": 0.1663736566901207,
      "num_tokens": 32724601.0,
      "step": 15765
    },
    {
      "entropy": 5.7427479267120365,
      "epoch": 1.4305152394775036,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004800045761216289,
      "loss": 5.4278,
      "mean_token_accuracy": 0.1658266693353653,
      "num_tokens": 32735015.0,
      "step": 15770
    },
    {
      "entropy": 5.7516786575317385,
      "epoch": 1.4309687953555879,
      "grad_norm": 2.21875,
      "learning_rate": 0.000479991239408135,
      "loss": 5.5416,
      "mean_token_accuracy": 0.1669219583272934,
      "num_tokens": 32745899.0,
      "step": 15775
    },
    {
      "entropy": 5.737471723556519,
      "epoch": 1.431422351233672,
      "grad_norm": 3.5,
      "learning_rate": 0.00047997789845537003,
      "loss": 5.4691,
      "mean_token_accuracy": 0.16777296215295792,
      "num_tokens": 32756052.0,
      "step": 15780
    },
    {
      "entropy": 5.702454900741577,
      "epoch": 1.4318759071117562,
      "grad_norm": 2.375,
      "learning_rate": 0.0004799645532636098,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16407757103443146,
      "num_tokens": 32767118.0,
      "step": 15785
    },
    {
      "entropy": 5.69314546585083,
      "epoch": 1.4323294629898404,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047995120383313044,
      "loss": 5.3612,
      "mean_token_accuracy": 0.17322630882263185,
      "num_tokens": 32777069.0,
      "step": 15790
    },
    {
      "entropy": 5.729848384857178,
      "epoch": 1.4327830188679245,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047993785016420796,
      "loss": 5.583,
      "mean_token_accuracy": 0.15602277368307113,
      "num_tokens": 32786967.0,
      "step": 15795
    },
    {
      "entropy": 5.7222531795501705,
      "epoch": 1.4332365747460087,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047992449225711854,
      "loss": 5.4535,
      "mean_token_accuracy": 0.16596826463937758,
      "num_tokens": 32797563.0,
      "step": 15800
    },
    {
      "entropy": 5.746530771255493,
      "epoch": 1.4336901306240928,
      "grad_norm": 3.125,
      "learning_rate": 0.00047991113011213837,
      "loss": 5.4437,
      "mean_token_accuracy": 0.16373340785503387,
      "num_tokens": 32808730.0,
      "step": 15805
    },
    {
      "entropy": 5.695686340332031,
      "epoch": 1.434143686502177,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004798977637295438,
      "loss": 5.4439,
      "mean_token_accuracy": 0.1689109355211258,
      "num_tokens": 32818050.0,
      "step": 15810
    },
    {
      "entropy": 5.704468870162964,
      "epoch": 1.4345972423802613,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004798843931096113,
      "loss": 5.4902,
      "mean_token_accuracy": 0.16171119660139083,
      "num_tokens": 32828993.0,
      "step": 15815
    },
    {
      "entropy": 5.7618875980377195,
      "epoch": 1.4350507982583454,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004798710182526173,
      "loss": 5.4819,
      "mean_token_accuracy": 0.17180464714765548,
      "num_tokens": 32839331.0,
      "step": 15820
    },
    {
      "entropy": 5.710322666168213,
      "epoch": 1.4355043541364296,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047985763915883847,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16102961003780364,
      "num_tokens": 32849946.0,
      "step": 15825
    },
    {
      "entropy": 5.678772687911987,
      "epoch": 1.4359579100145137,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004798442558285514,
      "loss": 5.4181,
      "mean_token_accuracy": 0.16813265085220336,
      "num_tokens": 32859960.0,
      "step": 15830
    },
    {
      "entropy": 5.751317644119263,
      "epoch": 1.436411465892598,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004798308682620329,
      "loss": 5.4486,
      "mean_token_accuracy": 0.16303409039974212,
      "num_tokens": 32870317.0,
      "step": 15835
    },
    {
      "entropy": 5.826468992233276,
      "epoch": 1.4368650217706822,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047981747645955996,
      "loss": 5.6147,
      "mean_token_accuracy": 0.15980882793664933,
      "num_tokens": 32880906.0,
      "step": 15840
    },
    {
      "entropy": 5.8070167064666744,
      "epoch": 1.4373185776487662,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004798040804214094,
      "loss": 5.634,
      "mean_token_accuracy": 0.16352577060461043,
      "num_tokens": 32891038.0,
      "step": 15845
    },
    {
      "entropy": 5.836701393127441,
      "epoch": 1.4377721335268505,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004797906801478582,
      "loss": 5.5951,
      "mean_token_accuracy": 0.1647000640630722,
      "num_tokens": 32901013.0,
      "step": 15850
    },
    {
      "entropy": 5.828676080703735,
      "epoch": 1.4382256894049346,
      "grad_norm": 15.25,
      "learning_rate": 0.0004797772756391836,
      "loss": 5.4962,
      "mean_token_accuracy": 0.16781063377857208,
      "num_tokens": 32911592.0,
      "step": 15855
    },
    {
      "entropy": 5.68819489479065,
      "epoch": 1.4386792452830188,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004797638668956627,
      "loss": 5.4392,
      "mean_token_accuracy": 0.1671498864889145,
      "num_tokens": 32922087.0,
      "step": 15860
    },
    {
      "entropy": 5.701917266845703,
      "epoch": 1.439132801161103,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004797504539175729,
      "loss": 5.5037,
      "mean_token_accuracy": 0.17037854790687562,
      "num_tokens": 32932039.0,
      "step": 15865
    },
    {
      "entropy": 5.749525022506714,
      "epoch": 1.4395863570391874,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004797370367051915,
      "loss": 5.4873,
      "mean_token_accuracy": 0.16582559198141097,
      "num_tokens": 32941896.0,
      "step": 15870
    },
    {
      "entropy": 5.761601448059082,
      "epoch": 1.4400399129172714,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000479723615258796,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1549222871661186,
      "num_tokens": 32951675.0,
      "step": 15875
    },
    {
      "entropy": 5.803951549530029,
      "epoch": 1.4404934687953557,
      "grad_norm": 3.171875,
      "learning_rate": 0.00047971018957866396,
      "loss": 5.5524,
      "mean_token_accuracy": 0.1685437649488449,
      "num_tokens": 32961378.0,
      "step": 15880
    },
    {
      "entropy": 5.821504306793213,
      "epoch": 1.4409470246734397,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047969675966507314,
      "loss": 5.4588,
      "mean_token_accuracy": 0.16860214322805406,
      "num_tokens": 32971770.0,
      "step": 15885
    },
    {
      "entropy": 5.756403160095215,
      "epoch": 1.441400580551524,
      "grad_norm": 4.125,
      "learning_rate": 0.000479683325518301,
      "loss": 5.4947,
      "mean_token_accuracy": 0.16291276961565018,
      "num_tokens": 32982934.0,
      "step": 15890
    },
    {
      "entropy": 5.689206933975219,
      "epoch": 1.4418541364296082,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004796698871386257,
      "loss": 5.4974,
      "mean_token_accuracy": 0.1626864567399025,
      "num_tokens": 32993252.0,
      "step": 15895
    },
    {
      "entropy": 5.769855880737305,
      "epoch": 1.4423076923076923,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004796564445263249,
      "loss": 5.5221,
      "mean_token_accuracy": 0.15902749300003052,
      "num_tokens": 33003610.0,
      "step": 15900
    },
    {
      "entropy": 5.805342817306519,
      "epoch": 1.4427612481857766,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047964299768167665,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16787874102592468,
      "num_tokens": 33014037.0,
      "step": 15905
    },
    {
      "entropy": 5.744533061981201,
      "epoch": 1.4432148040638606,
      "grad_norm": 3.25,
      "learning_rate": 0.0004796295466049591,
      "loss": 5.4977,
      "mean_token_accuracy": 0.16240752041339873,
      "num_tokens": 33023299.0,
      "step": 15910
    },
    {
      "entropy": 5.787428665161133,
      "epoch": 1.4436683599419449,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047961609129645036,
      "loss": 5.5266,
      "mean_token_accuracy": 0.16288958638906478,
      "num_tokens": 33034194.0,
      "step": 15915
    },
    {
      "entropy": 5.769773626327515,
      "epoch": 1.4441219158200291,
      "grad_norm": 3.296875,
      "learning_rate": 0.00047960263175642867,
      "loss": 5.419,
      "mean_token_accuracy": 0.16852306127548217,
      "num_tokens": 33044715.0,
      "step": 15920
    },
    {
      "entropy": 5.662240791320801,
      "epoch": 1.4445754716981132,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004795891679851724,
      "loss": 5.4687,
      "mean_token_accuracy": 0.16907748430967331,
      "num_tokens": 33055214.0,
      "step": 15925
    },
    {
      "entropy": 5.724631595611572,
      "epoch": 1.4450290275761974,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004795756999829601,
      "loss": 5.4646,
      "mean_token_accuracy": 0.16719101518392562,
      "num_tokens": 33065026.0,
      "step": 15930
    },
    {
      "entropy": 5.855519199371338,
      "epoch": 1.4454825834542815,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047956222775007016,
      "loss": 5.5321,
      "mean_token_accuracy": 0.16227805763483047,
      "num_tokens": 33075138.0,
      "step": 15935
    },
    {
      "entropy": 5.795980739593506,
      "epoch": 1.4459361393323658,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004795487512867812,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16377680748701096,
      "num_tokens": 33086625.0,
      "step": 15940
    },
    {
      "entropy": 5.798945140838623,
      "epoch": 1.44638969521045,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047953527059337194,
      "loss": 5.5686,
      "mean_token_accuracy": 0.15463654696941376,
      "num_tokens": 33097076.0,
      "step": 15945
    },
    {
      "entropy": 5.759383487701416,
      "epoch": 1.446843251088534,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004795217856701212,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16345080286264418,
      "num_tokens": 33107168.0,
      "step": 15950
    },
    {
      "entropy": 5.850946235656738,
      "epoch": 1.4472968069666183,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047950829651730774,
      "loss": 5.5477,
      "mean_token_accuracy": 0.16095320582389833,
      "num_tokens": 33117862.0,
      "step": 15955
    },
    {
      "entropy": 5.750865077972412,
      "epoch": 1.4477503628447024,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004794948031352107,
      "loss": 5.4496,
      "mean_token_accuracy": 0.16873406916856765,
      "num_tokens": 33127144.0,
      "step": 15960
    },
    {
      "entropy": 5.792923069000244,
      "epoch": 1.4482039187227866,
      "grad_norm": 3.296875,
      "learning_rate": 0.00047948130552410897,
      "loss": 5.6207,
      "mean_token_accuracy": 0.15654149353504182,
      "num_tokens": 33139180.0,
      "step": 15965
    },
    {
      "entropy": 5.8306097984313965,
      "epoch": 1.448657474600871,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047946780368428175,
      "loss": 5.5024,
      "mean_token_accuracy": 0.16083553731441497,
      "num_tokens": 33149467.0,
      "step": 15970
    },
    {
      "entropy": 5.728173732757568,
      "epoch": 1.449111030478955,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004794542976160081,
      "loss": 5.3667,
      "mean_token_accuracy": 0.17681725174188614,
      "num_tokens": 33158752.0,
      "step": 15975
    },
    {
      "entropy": 5.677011394500733,
      "epoch": 1.4495645863570392,
      "grad_norm": 3.5625,
      "learning_rate": 0.00047944078731956765,
      "loss": 5.4356,
      "mean_token_accuracy": 0.16876951605081558,
      "num_tokens": 33168242.0,
      "step": 15980
    },
    {
      "entropy": 5.665351867675781,
      "epoch": 1.4500181422351233,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047942727279523957,
      "loss": 5.4394,
      "mean_token_accuracy": 0.16484152525663376,
      "num_tokens": 33179134.0,
      "step": 15985
    },
    {
      "entropy": 5.736866664886475,
      "epoch": 1.4504716981132075,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004794137540433034,
      "loss": 5.5276,
      "mean_token_accuracy": 0.16109095960855485,
      "num_tokens": 33188388.0,
      "step": 15990
    },
    {
      "entropy": 5.780998659133911,
      "epoch": 1.4509252539912918,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004794002310640386,
      "loss": 5.48,
      "mean_token_accuracy": 0.16251239329576492,
      "num_tokens": 33198750.0,
      "step": 15995
    },
    {
      "entropy": 5.868821907043457,
      "epoch": 1.4513788098693758,
      "grad_norm": 3.921875,
      "learning_rate": 0.00047938670385772503,
      "loss": 5.5811,
      "mean_token_accuracy": 0.15783510059118272,
      "num_tokens": 33209532.0,
      "step": 16000
    },
    {
      "entropy": 5.792492055892945,
      "epoch": 1.45183236574746,
      "grad_norm": 3.1875,
      "learning_rate": 0.00047937317242464244,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1643729493021965,
      "num_tokens": 33219675.0,
      "step": 16005
    },
    {
      "entropy": 5.8150262355804445,
      "epoch": 1.4522859216255442,
      "grad_norm": 2.625,
      "learning_rate": 0.0004793596367650704,
      "loss": 5.5441,
      "mean_token_accuracy": 0.1580369219183922,
      "num_tokens": 33230919.0,
      "step": 16010
    },
    {
      "entropy": 5.768016290664673,
      "epoch": 1.4527394775036284,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004793460968792891,
      "loss": 5.4743,
      "mean_token_accuracy": 0.17275648266077043,
      "num_tokens": 33241146.0,
      "step": 16015
    },
    {
      "entropy": 5.730678701400757,
      "epoch": 1.4531930333817127,
      "grad_norm": 3.375,
      "learning_rate": 0.00047933255276757836,
      "loss": 5.5203,
      "mean_token_accuracy": 0.16372535973787308,
      "num_tokens": 33251434.0,
      "step": 16020
    },
    {
      "entropy": 5.835902261734009,
      "epoch": 1.4536465892597967,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004793190044302184,
      "loss": 5.603,
      "mean_token_accuracy": 0.15456404387950898,
      "num_tokens": 33261868.0,
      "step": 16025
    },
    {
      "entropy": 5.796180486679077,
      "epoch": 1.454100145137881,
      "grad_norm": 4.0,
      "learning_rate": 0.00047930545186748936,
      "loss": 5.4119,
      "mean_token_accuracy": 0.16823661774396897,
      "num_tokens": 33271798.0,
      "step": 16030
    },
    {
      "entropy": 5.749340057373047,
      "epoch": 1.454553701015965,
      "grad_norm": 4.34375,
      "learning_rate": 0.00047929189507967153,
      "loss": 5.5311,
      "mean_token_accuracy": 0.16348017901182174,
      "num_tokens": 33282614.0,
      "step": 16035
    },
    {
      "entropy": 5.669864273071289,
      "epoch": 1.4550072568940493,
      "grad_norm": 3.59375,
      "learning_rate": 0.00047927833406704516,
      "loss": 5.4516,
      "mean_token_accuracy": 0.1657632902264595,
      "num_tokens": 33292906.0,
      "step": 16040
    },
    {
      "entropy": 5.76576337814331,
      "epoch": 1.4554608127721336,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004792647688298909,
      "loss": 5.4595,
      "mean_token_accuracy": 0.16404654532670976,
      "num_tokens": 33302720.0,
      "step": 16045
    },
    {
      "entropy": 5.73647723197937,
      "epoch": 1.4559143686502178,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004792511993684891,
      "loss": 5.4402,
      "mean_token_accuracy": 0.168837833404541,
      "num_tokens": 33312959.0,
      "step": 16050
    },
    {
      "entropy": 5.6947352409362795,
      "epoch": 1.4563679245283019,
      "grad_norm": 2.875,
      "learning_rate": 0.00047923762568312053,
      "loss": 5.4041,
      "mean_token_accuracy": 0.1691950485110283,
      "num_tokens": 33322737.0,
      "step": 16055
    },
    {
      "entropy": 5.6799036979675295,
      "epoch": 1.4568214804063861,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004792240477740657,
      "loss": 5.4337,
      "mean_token_accuracy": 0.16971255987882614,
      "num_tokens": 33333185.0,
      "step": 16060
    },
    {
      "entropy": 5.798622560501099,
      "epoch": 1.4572750362844702,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004792104656416056,
      "loss": 5.5261,
      "mean_token_accuracy": 0.16138669699430466,
      "num_tokens": 33343414.0,
      "step": 16065
    },
    {
      "entropy": 5.7495715618133545,
      "epoch": 1.4577285921625545,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047919687928602104,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16166507601737976,
      "num_tokens": 33352527.0,
      "step": 16070
    },
    {
      "entropy": 5.803735828399658,
      "epoch": 1.4581821480406387,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047918328870759293,
      "loss": 5.5335,
      "mean_token_accuracy": 0.16170766502618789,
      "num_tokens": 33363526.0,
      "step": 16075
    },
    {
      "entropy": 5.766389560699463,
      "epoch": 1.4586357039187228,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004791696939066024,
      "loss": 5.5614,
      "mean_token_accuracy": 0.16453123986721038,
      "num_tokens": 33374302.0,
      "step": 16080
    },
    {
      "entropy": 5.778056001663208,
      "epoch": 1.459089259796807,
      "grad_norm": 2.625,
      "learning_rate": 0.0004791560948833306,
      "loss": 5.512,
      "mean_token_accuracy": 0.16594958901405335,
      "num_tokens": 33384786.0,
      "step": 16085
    },
    {
      "entropy": 5.791030216217041,
      "epoch": 1.459542815674891,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004791424916380586,
      "loss": 5.4875,
      "mean_token_accuracy": 0.1704912856221199,
      "num_tokens": 33394464.0,
      "step": 16090
    },
    {
      "entropy": 5.6948384761810305,
      "epoch": 1.4599963715529753,
      "grad_norm": 2.96875,
      "learning_rate": 0.000479128884171068,
      "loss": 5.5161,
      "mean_token_accuracy": 0.1628938674926758,
      "num_tokens": 33405476.0,
      "step": 16095
    },
    {
      "entropy": 5.721107864379883,
      "epoch": 1.4604499274310596,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047911527248264003,
      "loss": 5.5268,
      "mean_token_accuracy": 0.16545647829771043,
      "num_tokens": 33415758.0,
      "step": 16100
    },
    {
      "entropy": 5.767136573791504,
      "epoch": 1.4609034833091437,
      "grad_norm": 2.96875,
      "learning_rate": 0.00047910165657305613,
      "loss": 5.4985,
      "mean_token_accuracy": 0.16401995867490768,
      "num_tokens": 33424845.0,
      "step": 16105
    },
    {
      "entropy": 5.816438436508179,
      "epoch": 1.461357039187228,
      "grad_norm": 3.53125,
      "learning_rate": 0.00047908803644259804,
      "loss": 5.5435,
      "mean_token_accuracy": 0.160434989631176,
      "num_tokens": 33435928.0,
      "step": 16110
    },
    {
      "entropy": 5.751256704330444,
      "epoch": 1.461810595065312,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047907441209154726,
      "loss": 5.436,
      "mean_token_accuracy": 0.16507307589054107,
      "num_tokens": 33447307.0,
      "step": 16115
    },
    {
      "entropy": 5.7260230541229244,
      "epoch": 1.4622641509433962,
      "grad_norm": 2.125,
      "learning_rate": 0.00047906078352018575,
      "loss": 5.498,
      "mean_token_accuracy": 0.1597752347588539,
      "num_tokens": 33458116.0,
      "step": 16120
    },
    {
      "entropy": 5.6382973194122314,
      "epoch": 1.4627177068214805,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004790471507287952,
      "loss": 5.3856,
      "mean_token_accuracy": 0.16718733459711074,
      "num_tokens": 33469065.0,
      "step": 16125
    },
    {
      "entropy": 5.797253704071045,
      "epoch": 1.4631712626995645,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047903351371765757,
      "loss": 5.4817,
      "mean_token_accuracy": 0.1676560252904892,
      "num_tokens": 33479118.0,
      "step": 16130
    },
    {
      "entropy": 5.802631664276123,
      "epoch": 1.4636248185776488,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047901987248705484,
      "loss": 5.5271,
      "mean_token_accuracy": 0.16029046475887299,
      "num_tokens": 33489642.0,
      "step": 16135
    },
    {
      "entropy": 5.710444545745849,
      "epoch": 1.4640783744557329,
      "grad_norm": 3.125,
      "learning_rate": 0.0004790062270372693,
      "loss": 5.5127,
      "mean_token_accuracy": 0.1599433168768883,
      "num_tokens": 33499771.0,
      "step": 16140
    },
    {
      "entropy": 5.693857622146607,
      "epoch": 1.4645319303338171,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004789925773685828,
      "loss": 5.497,
      "mean_token_accuracy": 0.1615534394979477,
      "num_tokens": 33509975.0,
      "step": 16145
    },
    {
      "entropy": 5.7862061023712155,
      "epoch": 1.4649854862119014,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004789789234812779,
      "loss": 5.5259,
      "mean_token_accuracy": 0.1601164311170578,
      "num_tokens": 33520017.0,
      "step": 16150
    },
    {
      "entropy": 5.821394729614258,
      "epoch": 1.4654390420899854,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047896526537563686,
      "loss": 5.5868,
      "mean_token_accuracy": 0.15747144743800162,
      "num_tokens": 33531977.0,
      "step": 16155
    },
    {
      "entropy": 5.739919757843017,
      "epoch": 1.4658925979680697,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047895160305194223,
      "loss": 5.5291,
      "mean_token_accuracy": 0.1660194903612137,
      "num_tokens": 33543660.0,
      "step": 16160
    },
    {
      "entropy": 5.74508318901062,
      "epoch": 1.4663461538461537,
      "grad_norm": 2.125,
      "learning_rate": 0.0004789379365104764,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1658339336514473,
      "num_tokens": 33553814.0,
      "step": 16165
    },
    {
      "entropy": 5.801773929595948,
      "epoch": 1.466799709724238,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004789242657515221,
      "loss": 5.5015,
      "mean_token_accuracy": 0.16877730339765548,
      "num_tokens": 33564250.0,
      "step": 16170
    },
    {
      "entropy": 5.758831787109375,
      "epoch": 1.4672532656023223,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004789105907753621,
      "loss": 5.511,
      "mean_token_accuracy": 0.16126448810100555,
      "num_tokens": 33574755.0,
      "step": 16175
    },
    {
      "entropy": 5.742495059967041,
      "epoch": 1.4677068214804063,
      "grad_norm": 3.28125,
      "learning_rate": 0.000478896911582279,
      "loss": 5.4221,
      "mean_token_accuracy": 0.16934792697429657,
      "num_tokens": 33585268.0,
      "step": 16180
    },
    {
      "entropy": 5.7729967594146725,
      "epoch": 1.4681603773584906,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004788832281725558,
      "loss": 5.5568,
      "mean_token_accuracy": 0.16387153416872025,
      "num_tokens": 33595753.0,
      "step": 16185
    },
    {
      "entropy": 5.721085500717163,
      "epoch": 1.4686139332365746,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004788695405464755,
      "loss": 5.4177,
      "mean_token_accuracy": 0.17004877775907518,
      "num_tokens": 33605270.0,
      "step": 16190
    },
    {
      "entropy": 5.747508001327515,
      "epoch": 1.469067489114659,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047885584870432127,
      "loss": 5.5248,
      "mean_token_accuracy": 0.162787789106369,
      "num_tokens": 33615096.0,
      "step": 16195
    },
    {
      "entropy": 5.7317627429962155,
      "epoch": 1.4695210449927432,
      "grad_norm": 2.375,
      "learning_rate": 0.00047884215264637606,
      "loss": 5.4165,
      "mean_token_accuracy": 0.16615581810474395,
      "num_tokens": 33624770.0,
      "step": 16200
    },
    {
      "entropy": 5.7999255657196045,
      "epoch": 1.4699746008708272,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004788284523729232,
      "loss": 5.5392,
      "mean_token_accuracy": 0.15882084518671036,
      "num_tokens": 33634545.0,
      "step": 16205
    },
    {
      "entropy": 5.805088376998901,
      "epoch": 1.4704281567489115,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047881474788424595,
      "loss": 5.5302,
      "mean_token_accuracy": 0.16418834179639816,
      "num_tokens": 33644950.0,
      "step": 16210
    },
    {
      "entropy": 5.7786149978637695,
      "epoch": 1.4708817126269955,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047880103918062776,
      "loss": 5.4473,
      "mean_token_accuracy": 0.16866310238838195,
      "num_tokens": 33654369.0,
      "step": 16215
    },
    {
      "entropy": 5.653228998184204,
      "epoch": 1.4713352685050798,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047878732626235225,
      "loss": 5.483,
      "mean_token_accuracy": 0.16503912657499314,
      "num_tokens": 33664687.0,
      "step": 16220
    },
    {
      "entropy": 5.727343988418579,
      "epoch": 1.471788824383164,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004787736091297028,
      "loss": 5.459,
      "mean_token_accuracy": 0.17531538307666777,
      "num_tokens": 33675832.0,
      "step": 16225
    },
    {
      "entropy": 5.806529569625854,
      "epoch": 1.4722423802612483,
      "grad_norm": 4.0,
      "learning_rate": 0.0004787598877829633,
      "loss": 5.5363,
      "mean_token_accuracy": 0.16203751266002656,
      "num_tokens": 33686788.0,
      "step": 16230
    },
    {
      "entropy": 5.715418004989624,
      "epoch": 1.4726959361393324,
      "grad_norm": 2.125,
      "learning_rate": 0.0004787461622224173,
      "loss": 5.4452,
      "mean_token_accuracy": 0.16746071279048919,
      "num_tokens": 33696384.0,
      "step": 16235
    },
    {
      "entropy": 5.69289608001709,
      "epoch": 1.4731494920174166,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004787324324483488,
      "loss": 5.4738,
      "mean_token_accuracy": 0.16568130254745483,
      "num_tokens": 33707758.0,
      "step": 16240
    },
    {
      "entropy": 5.762115430831909,
      "epoch": 1.4736030478955007,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004787186984610416,
      "loss": 5.4522,
      "mean_token_accuracy": 0.17035746276378633,
      "num_tokens": 33717747.0,
      "step": 16245
    },
    {
      "entropy": 5.734479188919067,
      "epoch": 1.474056603773585,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047870496026077984,
      "loss": 5.4442,
      "mean_token_accuracy": 0.16524007469415664,
      "num_tokens": 33727960.0,
      "step": 16250
    },
    {
      "entropy": 5.668260335922241,
      "epoch": 1.4745101596516692,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004786912178478476,
      "loss": 5.4109,
      "mean_token_accuracy": 0.17311524450778962,
      "num_tokens": 33738874.0,
      "step": 16255
    },
    {
      "entropy": 5.719773006439209,
      "epoch": 1.4749637155297532,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004786774712225291,
      "loss": 5.433,
      "mean_token_accuracy": 0.17064379155635834,
      "num_tokens": 33749071.0,
      "step": 16260
    },
    {
      "entropy": 5.6876842975616455,
      "epoch": 1.4754172714078375,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004786637203851086,
      "loss": 5.4683,
      "mean_token_accuracy": 0.16505555510520936,
      "num_tokens": 33759509.0,
      "step": 16265
    },
    {
      "entropy": 5.79937858581543,
      "epoch": 1.4758708272859216,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004786499653358705,
      "loss": 5.5136,
      "mean_token_accuracy": 0.16523867398500441,
      "num_tokens": 33769974.0,
      "step": 16270
    },
    {
      "entropy": 5.850310754776001,
      "epoch": 1.4763243831640058,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004786362060750991,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16416485011577606,
      "num_tokens": 33780403.0,
      "step": 16275
    },
    {
      "entropy": 5.787898826599121,
      "epoch": 1.47677793904209,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047862244260307913,
      "loss": 5.5705,
      "mean_token_accuracy": 0.16401051878929138,
      "num_tokens": 33791364.0,
      "step": 16280
    },
    {
      "entropy": 5.7087541103363035,
      "epoch": 1.4772314949201741,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004786086749200951,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1665249913930893,
      "num_tokens": 33801964.0,
      "step": 16285
    },
    {
      "entropy": 5.776039981842041,
      "epoch": 1.4776850507982584,
      "grad_norm": 6.90625,
      "learning_rate": 0.0004785949030264319,
      "loss": 5.5063,
      "mean_token_accuracy": 0.16992122083902358,
      "num_tokens": 33813132.0,
      "step": 16290
    },
    {
      "entropy": 5.780887413024902,
      "epoch": 1.4781386066763424,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004785811269223741,
      "loss": 5.4818,
      "mean_token_accuracy": 0.1712091460824013,
      "num_tokens": 33823069.0,
      "step": 16295
    },
    {
      "entropy": 5.7973826885223385,
      "epoch": 1.4785921625544267,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004785673466082068,
      "loss": 5.4588,
      "mean_token_accuracy": 0.1616237133741379,
      "num_tokens": 33832626.0,
      "step": 16300
    },
    {
      "entropy": 5.712182092666626,
      "epoch": 1.479045718432511,
      "grad_norm": 2.625,
      "learning_rate": 0.00047855356208421484,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1654633492231369,
      "num_tokens": 33843384.0,
      "step": 16305
    },
    {
      "entropy": 5.735333442687988,
      "epoch": 1.479499274310595,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004785397733506834,
      "loss": 5.452,
      "mean_token_accuracy": 0.1661679431796074,
      "num_tokens": 33854286.0,
      "step": 16310
    },
    {
      "entropy": 5.733467960357666,
      "epoch": 1.4799528301886793,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004785259804078976,
      "loss": 5.4082,
      "mean_token_accuracy": 0.16801900118589402,
      "num_tokens": 33863379.0,
      "step": 16315
    },
    {
      "entropy": 5.8489296436309814,
      "epoch": 1.4804063860667633,
      "grad_norm": 2.625,
      "learning_rate": 0.0004785121832561426,
      "loss": 5.6539,
      "mean_token_accuracy": 0.15480876564979554,
      "num_tokens": 33875641.0,
      "step": 16320
    },
    {
      "entropy": 5.741404867172241,
      "epoch": 1.4808599419448476,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047849838189570375,
      "loss": 5.4241,
      "mean_token_accuracy": 0.1723483681678772,
      "num_tokens": 33885023.0,
      "step": 16325
    },
    {
      "entropy": 5.789837217330932,
      "epoch": 1.4813134978229319,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004784845763268666,
      "loss": 5.5766,
      "mean_token_accuracy": 0.15695818811655043,
      "num_tokens": 33895562.0,
      "step": 16330
    },
    {
      "entropy": 5.7498678207397464,
      "epoch": 1.481767053701016,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004784707665499165,
      "loss": 5.5151,
      "mean_token_accuracy": 0.16471377462148667,
      "num_tokens": 33907095.0,
      "step": 16335
    },
    {
      "entropy": 5.70077075958252,
      "epoch": 1.4822206095791002,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004784569525651391,
      "loss": 5.3057,
      "mean_token_accuracy": 0.17639741450548171,
      "num_tokens": 33917475.0,
      "step": 16340
    },
    {
      "entropy": 5.718368434906006,
      "epoch": 1.4826741654571842,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047844313437282005,
      "loss": 5.493,
      "mean_token_accuracy": 0.1679731786251068,
      "num_tokens": 33927740.0,
      "step": 16345
    },
    {
      "entropy": 5.71826400756836,
      "epoch": 1.4831277213352685,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047842931197324525,
      "loss": 5.4172,
      "mean_token_accuracy": 0.16887269914150238,
      "num_tokens": 33937416.0,
      "step": 16350
    },
    {
      "entropy": 5.692115926742554,
      "epoch": 1.4835812772133528,
      "grad_norm": 5.65625,
      "learning_rate": 0.00047841548536670033,
      "loss": 5.3885,
      "mean_token_accuracy": 0.17224257290363312,
      "num_tokens": 33946968.0,
      "step": 16355
    },
    {
      "entropy": 5.784520435333252,
      "epoch": 1.4840348330914368,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047840165455347136,
      "loss": 5.5315,
      "mean_token_accuracy": 0.15698693841695785,
      "num_tokens": 33957877.0,
      "step": 16360
    },
    {
      "entropy": 5.800479364395142,
      "epoch": 1.484488388969521,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047838781953384436,
      "loss": 5.5126,
      "mean_token_accuracy": 0.1636873245239258,
      "num_tokens": 33967890.0,
      "step": 16365
    },
    {
      "entropy": 5.672793006896972,
      "epoch": 1.484941944847605,
      "grad_norm": 2.75,
      "learning_rate": 0.0004783739803081054,
      "loss": 5.4654,
      "mean_token_accuracy": 0.16864327043294908,
      "num_tokens": 33977117.0,
      "step": 16370
    },
    {
      "entropy": 5.7175750732421875,
      "epoch": 1.4853955007256894,
      "grad_norm": 3.5625,
      "learning_rate": 0.00047836013687654075,
      "loss": 5.5504,
      "mean_token_accuracy": 0.1613817885518074,
      "num_tokens": 33986943.0,
      "step": 16375
    },
    {
      "entropy": 5.716193437576294,
      "epoch": 1.4858490566037736,
      "grad_norm": 2.875,
      "learning_rate": 0.00047834628923943654,
      "loss": 5.4532,
      "mean_token_accuracy": 0.1662539705634117,
      "num_tokens": 33996499.0,
      "step": 16380
    },
    {
      "entropy": 5.805265045166015,
      "epoch": 1.4863026124818577,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004783324373970793,
      "loss": 5.5632,
      "mean_token_accuracy": 0.16451188176870346,
      "num_tokens": 34006749.0,
      "step": 16385
    },
    {
      "entropy": 5.866697072982788,
      "epoch": 1.486756168359942,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004783185813497555,
      "loss": 5.5806,
      "mean_token_accuracy": 0.155438344925642,
      "num_tokens": 34017194.0,
      "step": 16390
    },
    {
      "entropy": 5.807399034500122,
      "epoch": 1.487209724238026,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004783047210977516,
      "loss": 5.4958,
      "mean_token_accuracy": 0.16723082214593887,
      "num_tokens": 34027386.0,
      "step": 16395
    },
    {
      "entropy": 5.833196449279785,
      "epoch": 1.4876632801161103,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047829085664135424,
      "loss": 5.5412,
      "mean_token_accuracy": 0.16031222194433212,
      "num_tokens": 34037540.0,
      "step": 16400
    },
    {
      "entropy": 5.759831428527832,
      "epoch": 1.4881168359941945,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047827698798085015,
      "loss": 5.5018,
      "mean_token_accuracy": 0.16242261677980424,
      "num_tokens": 34048008.0,
      "step": 16405
    },
    {
      "entropy": 5.795193243026733,
      "epoch": 1.4885703918722788,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004782631151165262,
      "loss": 5.519,
      "mean_token_accuracy": 0.1645362123847008,
      "num_tokens": 34058009.0,
      "step": 16410
    },
    {
      "entropy": 5.817488718032837,
      "epoch": 1.4890239477503628,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047824923804866914,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1658645510673523,
      "num_tokens": 34067625.0,
      "step": 16415
    },
    {
      "entropy": 5.677306604385376,
      "epoch": 1.4894775036284469,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047823535677756607,
      "loss": 5.4585,
      "mean_token_accuracy": 0.16158679723739625,
      "num_tokens": 34077349.0,
      "step": 16420
    },
    {
      "entropy": 5.796026134490967,
      "epoch": 1.4899310595065312,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004782214713035041,
      "loss": 5.5894,
      "mean_token_accuracy": 0.159983092546463,
      "num_tokens": 34088636.0,
      "step": 16425
    },
    {
      "entropy": 5.797797536849975,
      "epoch": 1.4903846153846154,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047820758162677026,
      "loss": 5.5017,
      "mean_token_accuracy": 0.17131305038928984,
      "num_tokens": 34098806.0,
      "step": 16430
    },
    {
      "entropy": 5.759490156173706,
      "epoch": 1.4908381712626997,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047819368774765187,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16527411937713624,
      "num_tokens": 34109713.0,
      "step": 16435
    },
    {
      "entropy": 5.698564434051514,
      "epoch": 1.4912917271407837,
      "grad_norm": 3.828125,
      "learning_rate": 0.00047817978966643626,
      "loss": 5.4663,
      "mean_token_accuracy": 0.1692475214600563,
      "num_tokens": 34119585.0,
      "step": 16440
    },
    {
      "entropy": 5.855516958236694,
      "epoch": 1.491745283018868,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004781658873834108,
      "loss": 5.5064,
      "mean_token_accuracy": 0.15707526504993438,
      "num_tokens": 34131245.0,
      "step": 16445
    },
    {
      "entropy": 5.797071313858032,
      "epoch": 1.492198838896952,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047815198089886296,
      "loss": 5.4836,
      "mean_token_accuracy": 0.1690132051706314,
      "num_tokens": 34141229.0,
      "step": 16450
    },
    {
      "entropy": 5.725530958175659,
      "epoch": 1.4926523947750363,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004781380702130805,
      "loss": 5.5281,
      "mean_token_accuracy": 0.16326702237129212,
      "num_tokens": 34151781.0,
      "step": 16455
    },
    {
      "entropy": 5.770390224456787,
      "epoch": 1.4931059506531206,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047812415532635087,
      "loss": 5.4545,
      "mean_token_accuracy": 0.167418971657753,
      "num_tokens": 34162305.0,
      "step": 16460
    },
    {
      "entropy": 5.815627193450927,
      "epoch": 1.4935595065312046,
      "grad_norm": 3.328125,
      "learning_rate": 0.000478110236238962,
      "loss": 5.5734,
      "mean_token_accuracy": 0.16374657452106475,
      "num_tokens": 34173613.0,
      "step": 16465
    },
    {
      "entropy": 5.7491700649261475,
      "epoch": 1.4940130624092889,
      "grad_norm": 4.4375,
      "learning_rate": 0.00047809631295120164,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1616842418909073,
      "num_tokens": 34183540.0,
      "step": 16470
    },
    {
      "entropy": 5.7139934539794925,
      "epoch": 1.494466618287373,
      "grad_norm": 2.0,
      "learning_rate": 0.00047808238546335786,
      "loss": 5.4888,
      "mean_token_accuracy": 0.1623132586479187,
      "num_tokens": 34193292.0,
      "step": 16475
    },
    {
      "entropy": 5.778197002410889,
      "epoch": 1.4949201741654572,
      "grad_norm": 3.203125,
      "learning_rate": 0.00047806845377571856,
      "loss": 5.5293,
      "mean_token_accuracy": 0.1672927275300026,
      "num_tokens": 34202697.0,
      "step": 16480
    },
    {
      "entropy": 5.7389201641082765,
      "epoch": 1.4953737300435415,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047805451788857187,
      "loss": 5.4118,
      "mean_token_accuracy": 0.17393453121185304,
      "num_tokens": 34213396.0,
      "step": 16485
    },
    {
      "entropy": 5.785162353515625,
      "epoch": 1.4958272859216255,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000478040577802206,
      "loss": 5.4388,
      "mean_token_accuracy": 0.16598025411367417,
      "num_tokens": 34223020.0,
      "step": 16490
    },
    {
      "entropy": 5.717736911773682,
      "epoch": 1.4962808417997098,
      "grad_norm": 3.75,
      "learning_rate": 0.0004780266335169093,
      "loss": 5.4833,
      "mean_token_accuracy": 0.16676653921604156,
      "num_tokens": 34233249.0,
      "step": 16495
    },
    {
      "entropy": 5.7663366317749025,
      "epoch": 1.4967343976777938,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047801268503297,
      "loss": 5.5332,
      "mean_token_accuracy": 0.16241997629404067,
      "num_tokens": 34244091.0,
      "step": 16500
    },
    {
      "entropy": 5.75670018196106,
      "epoch": 1.497187953555878,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047799873235067667,
      "loss": 5.5179,
      "mean_token_accuracy": 0.16847553104162216,
      "num_tokens": 34255060.0,
      "step": 16505
    },
    {
      "entropy": 5.761952018737793,
      "epoch": 1.4976415094339623,
      "grad_norm": 2.5,
      "learning_rate": 0.0004779847754703178,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17746998220682145,
      "num_tokens": 34264361.0,
      "step": 16510
    },
    {
      "entropy": 5.732575988769531,
      "epoch": 1.4980950653120464,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004779708143921821,
      "loss": 5.4749,
      "mean_token_accuracy": 0.16618917286396026,
      "num_tokens": 34274480.0,
      "step": 16515
    },
    {
      "entropy": 5.706440734863281,
      "epoch": 1.4985486211901307,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047795684911655824,
      "loss": 5.497,
      "mean_token_accuracy": 0.16535889953374863,
      "num_tokens": 34285951.0,
      "step": 16520
    },
    {
      "entropy": 5.791449356079101,
      "epoch": 1.4990021770682147,
      "grad_norm": 2.40625,
      "learning_rate": 0.000477942879643735,
      "loss": 5.5303,
      "mean_token_accuracy": 0.160249263048172,
      "num_tokens": 34296384.0,
      "step": 16525
    },
    {
      "entropy": 5.747885799407959,
      "epoch": 1.499455732946299,
      "grad_norm": 2.5,
      "learning_rate": 0.00047792890597400125,
      "loss": 5.4743,
      "mean_token_accuracy": 0.16591938883066176,
      "num_tokens": 34306776.0,
      "step": 16530
    },
    {
      "entropy": 5.807723045349121,
      "epoch": 1.4999092888243832,
      "grad_norm": 2.25,
      "learning_rate": 0.00047791492810764604,
      "loss": 5.4697,
      "mean_token_accuracy": 0.16555147171020507,
      "num_tokens": 34317047.0,
      "step": 16535
    },
    {
      "entropy": 5.742052936553955,
      "epoch": 1.5003628447024675,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004779009460449584,
      "loss": 5.4197,
      "mean_token_accuracy": 0.16636399775743485,
      "num_tokens": 34326392.0,
      "step": 16540
    },
    {
      "entropy": 5.747958850860596,
      "epoch": 1.5008164005805515,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047788695978622756,
      "loss": 5.4721,
      "mean_token_accuracy": 0.17087629437446594,
      "num_tokens": 34336580.0,
      "step": 16545
    },
    {
      "entropy": 5.755423641204834,
      "epoch": 1.5012699564586356,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047787296933174265,
      "loss": 5.5104,
      "mean_token_accuracy": 0.16620604395866395,
      "num_tokens": 34346792.0,
      "step": 16550
    },
    {
      "entropy": 5.817895841598511,
      "epoch": 1.5017235123367199,
      "grad_norm": 2.828125,
      "learning_rate": 0.000477858974681793,
      "loss": 5.4968,
      "mean_token_accuracy": 0.1640177845954895,
      "num_tokens": 34356675.0,
      "step": 16555
    },
    {
      "entropy": 5.766955375671387,
      "epoch": 1.5021770682148041,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004778449758366681,
      "loss": 5.5084,
      "mean_token_accuracy": 0.16773485243320466,
      "num_tokens": 34366357.0,
      "step": 16560
    },
    {
      "entropy": 5.72684965133667,
      "epoch": 1.5026306240928884,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004778309727966574,
      "loss": 5.4863,
      "mean_token_accuracy": 0.16593327522277831,
      "num_tokens": 34377229.0,
      "step": 16565
    },
    {
      "entropy": 5.672737073898316,
      "epoch": 1.5030841799709724,
      "grad_norm": 3.0,
      "learning_rate": 0.00047781696556205045,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16962171196937562,
      "num_tokens": 34387873.0,
      "step": 16570
    },
    {
      "entropy": 5.781691694259644,
      "epoch": 1.5035377358490565,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047780295413313696,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1718684747815132,
      "num_tokens": 34398889.0,
      "step": 16575
    },
    {
      "entropy": 5.730648517608643,
      "epoch": 1.5039912917271407,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047778893851020676,
      "loss": 5.4117,
      "mean_token_accuracy": 0.1692804366350174,
      "num_tokens": 34408508.0,
      "step": 16580
    },
    {
      "entropy": 5.696586656570434,
      "epoch": 1.504444847605225,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004777749186935496,
      "loss": 5.5478,
      "mean_token_accuracy": 0.1637800857424736,
      "num_tokens": 34418928.0,
      "step": 16585
    },
    {
      "entropy": 5.752023553848266,
      "epoch": 1.5048984034833093,
      "grad_norm": 2.5,
      "learning_rate": 0.0004777608946834555,
      "loss": 5.5237,
      "mean_token_accuracy": 0.1610870987176895,
      "num_tokens": 34429789.0,
      "step": 16590
    },
    {
      "entropy": 5.815820693969727,
      "epoch": 1.5053519593613933,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004777468664802144,
      "loss": 5.6002,
      "mean_token_accuracy": 0.15755764096975328,
      "num_tokens": 34440795.0,
      "step": 16595
    },
    {
      "entropy": 5.826911163330078,
      "epoch": 1.5058055152394774,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047773283408411636,
      "loss": 5.5039,
      "mean_token_accuracy": 0.15856851264834404,
      "num_tokens": 34450772.0,
      "step": 16600
    },
    {
      "entropy": 5.718824768066407,
      "epoch": 1.5062590711175616,
      "grad_norm": 3.15625,
      "learning_rate": 0.00047771879749545175,
      "loss": 5.4414,
      "mean_token_accuracy": 0.16626231223344803,
      "num_tokens": 34461273.0,
      "step": 16605
    },
    {
      "entropy": 5.7022199630737305,
      "epoch": 1.506712626995646,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047770475671451073,
      "loss": 5.5154,
      "mean_token_accuracy": 0.16265685111284256,
      "num_tokens": 34470942.0,
      "step": 16610
    },
    {
      "entropy": 5.763551998138428,
      "epoch": 1.5071661828737302,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004776907117415837,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16591255664825438,
      "num_tokens": 34481117.0,
      "step": 16615
    },
    {
      "entropy": 5.722925996780395,
      "epoch": 1.5076197387518142,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047767666257696103,
      "loss": 5.4308,
      "mean_token_accuracy": 0.16904795318841934,
      "num_tokens": 34490953.0,
      "step": 16620
    },
    {
      "entropy": 5.764816808700561,
      "epoch": 1.5080732946298983,
      "grad_norm": 4.5625,
      "learning_rate": 0.00047766260922093337,
      "loss": 5.3953,
      "mean_token_accuracy": 0.17081756442785262,
      "num_tokens": 34502765.0,
      "step": 16625
    },
    {
      "entropy": 5.860283660888672,
      "epoch": 1.5085268505079825,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004776485516737913,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15801553651690484,
      "num_tokens": 34514473.0,
      "step": 16630
    },
    {
      "entropy": 5.851029253005981,
      "epoch": 1.5089804063860668,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004776344899358256,
      "loss": 5.5847,
      "mean_token_accuracy": 0.1599692076444626,
      "num_tokens": 34524218.0,
      "step": 16635
    },
    {
      "entropy": 5.7593060493469235,
      "epoch": 1.509433962264151,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004776204240073269,
      "loss": 5.4906,
      "mean_token_accuracy": 0.16537062227725982,
      "num_tokens": 34535697.0,
      "step": 16640
    },
    {
      "entropy": 5.751308822631836,
      "epoch": 1.509887518142235,
      "grad_norm": 3.0,
      "learning_rate": 0.00047760635388858624,
      "loss": 5.4989,
      "mean_token_accuracy": 0.17202635407447814,
      "num_tokens": 34545130.0,
      "step": 16645
    },
    {
      "entropy": 5.707167673110962,
      "epoch": 1.5103410740203191,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047759227957989465,
      "loss": 5.4454,
      "mean_token_accuracy": 0.1687132552266121,
      "num_tokens": 34554974.0,
      "step": 16650
    },
    {
      "entropy": 5.76544599533081,
      "epoch": 1.5107946298984034,
      "grad_norm": 4.09375,
      "learning_rate": 0.000477578201081543,
      "loss": 5.4866,
      "mean_token_accuracy": 0.16432657539844514,
      "num_tokens": 34565660.0,
      "step": 16655
    },
    {
      "entropy": 5.775887775421142,
      "epoch": 1.5112481857764877,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004775641183938225,
      "loss": 5.5116,
      "mean_token_accuracy": 0.16381104439496993,
      "num_tokens": 34576447.0,
      "step": 16660
    },
    {
      "entropy": 5.829549837112427,
      "epoch": 1.511701741654572,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047755003151702443,
      "loss": 5.6495,
      "mean_token_accuracy": 0.15984634608030318,
      "num_tokens": 34588333.0,
      "step": 16665
    },
    {
      "entropy": 5.7317101001739506,
      "epoch": 1.512155297532656,
      "grad_norm": 3.34375,
      "learning_rate": 0.00047753594045144015,
      "loss": 5.3872,
      "mean_token_accuracy": 0.1669989660382271,
      "num_tokens": 34598166.0,
      "step": 16670
    },
    {
      "entropy": 5.7106107711792,
      "epoch": 1.5126088534107403,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004775218451973609,
      "loss": 5.422,
      "mean_token_accuracy": 0.16638735830783843,
      "num_tokens": 34608615.0,
      "step": 16675
    },
    {
      "entropy": 5.741033840179443,
      "epoch": 1.5130624092888243,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004775077457550784,
      "loss": 5.487,
      "mean_token_accuracy": 0.1639235347509384,
      "num_tokens": 34618618.0,
      "step": 16680
    },
    {
      "entropy": 5.720541477203369,
      "epoch": 1.5135159651669086,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047749364212488403,
      "loss": 5.5143,
      "mean_token_accuracy": 0.16121886372566224,
      "num_tokens": 34629412.0,
      "step": 16685
    },
    {
      "entropy": 5.807133769989013,
      "epoch": 1.5139695210449928,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004774795343070696,
      "loss": 5.5735,
      "mean_token_accuracy": 0.16632983684539795,
      "num_tokens": 34640487.0,
      "step": 16690
    },
    {
      "entropy": 5.7874682426452635,
      "epoch": 1.5144230769230769,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004774654223019268,
      "loss": 5.4684,
      "mean_token_accuracy": 0.17152667939662933,
      "num_tokens": 34650126.0,
      "step": 16695
    },
    {
      "entropy": 5.693085718154907,
      "epoch": 1.5148766328011611,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004774513061097474,
      "loss": 5.3607,
      "mean_token_accuracy": 0.17249772548675538,
      "num_tokens": 34659721.0,
      "step": 16700
    },
    {
      "entropy": 5.677116394042969,
      "epoch": 1.5153301886792452,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004774371857308235,
      "loss": 5.4723,
      "mean_token_accuracy": 0.16534443497657775,
      "num_tokens": 34670611.0,
      "step": 16705
    },
    {
      "entropy": 5.732109785079956,
      "epoch": 1.5157837445573294,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047742306116544693,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1719784617424011,
      "num_tokens": 34680173.0,
      "step": 16710
    },
    {
      "entropy": 5.758274078369141,
      "epoch": 1.5162373004354137,
      "grad_norm": 6.78125,
      "learning_rate": 0.00047740893241390994,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16809773594141006,
      "num_tokens": 34690176.0,
      "step": 16715
    },
    {
      "entropy": 5.793875408172608,
      "epoch": 1.516690856313498,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004773947994765046,
      "loss": 5.4466,
      "mean_token_accuracy": 0.17031722515821457,
      "num_tokens": 34700481.0,
      "step": 16720
    },
    {
      "entropy": 5.762631845474243,
      "epoch": 1.517144412191582,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004773806623535232,
      "loss": 5.6247,
      "mean_token_accuracy": 0.16147315353155137,
      "num_tokens": 34710643.0,
      "step": 16725
    },
    {
      "entropy": 5.770838069915771,
      "epoch": 1.517597968069666,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047736652104525816,
      "loss": 5.484,
      "mean_token_accuracy": 0.16311268806457518,
      "num_tokens": 34720998.0,
      "step": 16730
    },
    {
      "entropy": 5.809461450576782,
      "epoch": 1.5180515239477503,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004773523755520019,
      "loss": 5.5495,
      "mean_token_accuracy": 0.15910750925540923,
      "num_tokens": 34731071.0,
      "step": 16735
    },
    {
      "entropy": 5.751518058776855,
      "epoch": 1.5185050798258346,
      "grad_norm": 4.5625,
      "learning_rate": 0.000477338225874047,
      "loss": 5.4147,
      "mean_token_accuracy": 0.1701589807868004,
      "num_tokens": 34741346.0,
      "step": 16740
    },
    {
      "entropy": 5.760993719100952,
      "epoch": 1.5189586357039189,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047732407201168595,
      "loss": 5.4331,
      "mean_token_accuracy": 0.16575501263141632,
      "num_tokens": 34751092.0,
      "step": 16745
    },
    {
      "entropy": 5.81353063583374,
      "epoch": 1.519412191582003,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004773099139652116,
      "loss": 5.5577,
      "mean_token_accuracy": 0.16306657195091248,
      "num_tokens": 34761477.0,
      "step": 16750
    },
    {
      "entropy": 5.6945152759552,
      "epoch": 1.519865747460087,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004772957517349167,
      "loss": 5.4799,
      "mean_token_accuracy": 0.16661242097616197,
      "num_tokens": 34772491.0,
      "step": 16755
    },
    {
      "entropy": 5.723292875289917,
      "epoch": 1.5203193033381712,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004772815853210939,
      "loss": 5.4933,
      "mean_token_accuracy": 0.16406753063201904,
      "num_tokens": 34783192.0,
      "step": 16760
    },
    {
      "entropy": 5.750239372253418,
      "epoch": 1.5207728592162555,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047726741472403655,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16709423512220384,
      "num_tokens": 34794164.0,
      "step": 16765
    },
    {
      "entropy": 5.742389869689942,
      "epoch": 1.5212264150943398,
      "grad_norm": 3.140625,
      "learning_rate": 0.00047725323994403753,
      "loss": 5.5145,
      "mean_token_accuracy": 0.1550010234117508,
      "num_tokens": 34804390.0,
      "step": 16770
    },
    {
      "entropy": 5.875350522994995,
      "epoch": 1.5216799709724238,
      "grad_norm": 4.5625,
      "learning_rate": 0.00047723906098138993,
      "loss": 5.6148,
      "mean_token_accuracy": 0.15728134661912918,
      "num_tokens": 34814751.0,
      "step": 16775
    },
    {
      "entropy": 5.8044762134552,
      "epoch": 1.5221335268505078,
      "grad_norm": 3.65625,
      "learning_rate": 0.00047722487783638697,
      "loss": 5.4597,
      "mean_token_accuracy": 0.17191277742385863,
      "num_tokens": 34824977.0,
      "step": 16780
    },
    {
      "entropy": 5.7247716903686525,
      "epoch": 1.5225870827285921,
      "grad_norm": 3.171875,
      "learning_rate": 0.000477210690509322,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16733710318803788,
      "num_tokens": 34835540.0,
      "step": 16785
    },
    {
      "entropy": 5.747857284545899,
      "epoch": 1.5230406386066764,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047719649900048843,
      "loss": 5.4713,
      "mean_token_accuracy": 0.1621651217341423,
      "num_tokens": 34845233.0,
      "step": 16790
    },
    {
      "entropy": 5.869965839385986,
      "epoch": 1.5234941944847606,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047718230331017975,
      "loss": 5.5843,
      "mean_token_accuracy": 0.16166191622614862,
      "num_tokens": 34855756.0,
      "step": 16795
    },
    {
      "entropy": 5.804520750045777,
      "epoch": 1.5239477503628447,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047716810343868956,
      "loss": 5.4974,
      "mean_token_accuracy": 0.15590585619211197,
      "num_tokens": 34867869.0,
      "step": 16800
    },
    {
      "entropy": 5.795978212356568,
      "epoch": 1.5244013062409287,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047715389938631143,
      "loss": 5.5622,
      "mean_token_accuracy": 0.16301123499870301,
      "num_tokens": 34878625.0,
      "step": 16805
    },
    {
      "entropy": 5.828445053100586,
      "epoch": 1.524854862119013,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047713969115333916,
      "loss": 5.5176,
      "mean_token_accuracy": 0.15993245542049409,
      "num_tokens": 34889408.0,
      "step": 16810
    },
    {
      "entropy": 5.792006921768189,
      "epoch": 1.5253084179970973,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004771254787400666,
      "loss": 5.5769,
      "mean_token_accuracy": 0.1568712681531906,
      "num_tokens": 34899602.0,
      "step": 16815
    },
    {
      "entropy": 5.724984502792358,
      "epoch": 1.5257619738751815,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004771112621467876,
      "loss": 5.4041,
      "mean_token_accuracy": 0.16784977167844772,
      "num_tokens": 34910215.0,
      "step": 16820
    },
    {
      "entropy": 5.73494143486023,
      "epoch": 1.5262155297532656,
      "grad_norm": 4.0,
      "learning_rate": 0.0004770970413737961,
      "loss": 5.478,
      "mean_token_accuracy": 0.16947001963853836,
      "num_tokens": 34919641.0,
      "step": 16825
    },
    {
      "entropy": 5.744553899765014,
      "epoch": 1.5266690856313496,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004770828164213865,
      "loss": 5.3938,
      "mean_token_accuracy": 0.17306017279624938,
      "num_tokens": 34930243.0,
      "step": 16830
    },
    {
      "entropy": 5.760212755203247,
      "epoch": 1.5271226415094339,
      "grad_norm": 3.875,
      "learning_rate": 0.0004770685872898525,
      "loss": 5.5421,
      "mean_token_accuracy": 0.1675536960363388,
      "num_tokens": 34941057.0,
      "step": 16835
    },
    {
      "entropy": 5.785358953475952,
      "epoch": 1.5275761973875182,
      "grad_norm": 2.5,
      "learning_rate": 0.00047705435397948876,
      "loss": 5.4058,
      "mean_token_accuracy": 0.16808340847492217,
      "num_tokens": 34951150.0,
      "step": 16840
    },
    {
      "entropy": 5.80018162727356,
      "epoch": 1.5280297532656024,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004770401164905895,
      "loss": 5.5164,
      "mean_token_accuracy": 0.15894629806280136,
      "num_tokens": 34961648.0,
      "step": 16845
    },
    {
      "entropy": 5.749872064590454,
      "epoch": 1.5284833091436865,
      "grad_norm": 3.5,
      "learning_rate": 0.00047702587482344915,
      "loss": 5.5118,
      "mean_token_accuracy": 0.16121797859668732,
      "num_tokens": 34971573.0,
      "step": 16850
    },
    {
      "entropy": 5.760915279388428,
      "epoch": 1.5289368650217707,
      "grad_norm": 2.625,
      "learning_rate": 0.0004770116289783622,
      "loss": 5.4954,
      "mean_token_accuracy": 0.16071234345436097,
      "num_tokens": 34981984.0,
      "step": 16855
    },
    {
      "entropy": 5.759790897369385,
      "epoch": 1.5293904208998548,
      "grad_norm": 4.0,
      "learning_rate": 0.0004769973789556233,
      "loss": 5.4905,
      "mean_token_accuracy": 0.1621873825788498,
      "num_tokens": 34990861.0,
      "step": 16860
    },
    {
      "entropy": 5.763146781921387,
      "epoch": 1.529843976777939,
      "grad_norm": 4.0625,
      "learning_rate": 0.00047698312475552714,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16173181533813477,
      "num_tokens": 35001504.0,
      "step": 16865
    },
    {
      "entropy": 5.796394777297974,
      "epoch": 1.5302975326560233,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004769688663783684,
      "loss": 5.4436,
      "mean_token_accuracy": 0.16866615563631057,
      "num_tokens": 35012425.0,
      "step": 16870
    },
    {
      "entropy": 5.770410251617432,
      "epoch": 1.5307510885341074,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004769546038244421,
      "loss": 5.4265,
      "mean_token_accuracy": 0.16738052517175675,
      "num_tokens": 35022854.0,
      "step": 16875
    },
    {
      "entropy": 5.7480902671813965,
      "epoch": 1.5312046444121916,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047694033709404313,
      "loss": 5.531,
      "mean_token_accuracy": 0.16326771229505538,
      "num_tokens": 35032694.0,
      "step": 16880
    },
    {
      "entropy": 5.73104453086853,
      "epoch": 1.5316582002902757,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047692606618746655,
      "loss": 5.4407,
      "mean_token_accuracy": 0.16918860673904418,
      "num_tokens": 35042217.0,
      "step": 16885
    },
    {
      "entropy": 5.766095113754273,
      "epoch": 1.53211175616836,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047691179110500735,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1574793577194214,
      "num_tokens": 35052831.0,
      "step": 16890
    },
    {
      "entropy": 5.8288397789001465,
      "epoch": 1.5325653120464442,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047689751184696097,
      "loss": 5.5609,
      "mean_token_accuracy": 0.16141264587640763,
      "num_tokens": 35061955.0,
      "step": 16895
    },
    {
      "entropy": 5.78143539428711,
      "epoch": 1.5330188679245285,
      "grad_norm": 2.75,
      "learning_rate": 0.0004768832284136226,
      "loss": 5.4671,
      "mean_token_accuracy": 0.16387688368558884,
      "num_tokens": 35072526.0,
      "step": 16900
    },
    {
      "entropy": 5.729931259155274,
      "epoch": 1.5334724238026125,
      "grad_norm": 4.25,
      "learning_rate": 0.0004768689408052876,
      "loss": 5.4205,
      "mean_token_accuracy": 0.1709127888083458,
      "num_tokens": 35082415.0,
      "step": 16905
    },
    {
      "entropy": 5.722542953491211,
      "epoch": 1.5339259796806965,
      "grad_norm": 3.46875,
      "learning_rate": 0.00047685464902225144,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16714812815189362,
      "num_tokens": 35092805.0,
      "step": 16910
    },
    {
      "entropy": 5.760873079299927,
      "epoch": 1.5343795355587808,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004768403530648097,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16846604347229005,
      "num_tokens": 35102963.0,
      "step": 16915
    },
    {
      "entropy": 5.823122167587281,
      "epoch": 1.534833091436865,
      "grad_norm": 3.8125,
      "learning_rate": 0.00047682605293325806,
      "loss": 5.5145,
      "mean_token_accuracy": 0.16570364832878112,
      "num_tokens": 35114140.0,
      "step": 16920
    },
    {
      "entropy": 5.840953969955445,
      "epoch": 1.5352866473149493,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047681174862789215,
      "loss": 5.5722,
      "mean_token_accuracy": 0.16135205179452897,
      "num_tokens": 35125314.0,
      "step": 16925
    },
    {
      "entropy": 5.771080493927002,
      "epoch": 1.5357402031930334,
      "grad_norm": 4.375,
      "learning_rate": 0.00047679744014900795,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16519787460565566,
      "num_tokens": 35134934.0,
      "step": 16930
    },
    {
      "entropy": 5.720629644393921,
      "epoch": 1.5361937590711174,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004767831274969011,
      "loss": 5.4361,
      "mean_token_accuracy": 0.1688353344798088,
      "num_tokens": 35145341.0,
      "step": 16935
    },
    {
      "entropy": 5.718237209320068,
      "epoch": 1.5366473149492017,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047676881067186793,
      "loss": 5.4111,
      "mean_token_accuracy": 0.1650972455739975,
      "num_tokens": 35156046.0,
      "step": 16940
    },
    {
      "entropy": 5.775576639175415,
      "epoch": 1.537100870827286,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047675448967420426,
      "loss": 5.5756,
      "mean_token_accuracy": 0.16162624210119247,
      "num_tokens": 35168232.0,
      "step": 16945
    },
    {
      "entropy": 5.836013460159302,
      "epoch": 1.5375544267053702,
      "grad_norm": 3.375,
      "learning_rate": 0.0004767401645042064,
      "loss": 5.5367,
      "mean_token_accuracy": 0.1628153592348099,
      "num_tokens": 35178585.0,
      "step": 16950
    },
    {
      "entropy": 5.729782772064209,
      "epoch": 1.5380079825834543,
      "grad_norm": 3.0,
      "learning_rate": 0.00047672583516217043,
      "loss": 5.4091,
      "mean_token_accuracy": 0.16760532408952714,
      "num_tokens": 35189417.0,
      "step": 16955
    },
    {
      "entropy": 5.715841245651245,
      "epoch": 1.5384615384615383,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004767115016483928,
      "loss": 5.4244,
      "mean_token_accuracy": 0.17102244794368743,
      "num_tokens": 35200167.0,
      "step": 16960
    },
    {
      "entropy": 5.718662881851197,
      "epoch": 1.5389150943396226,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004766971639631699,
      "loss": 5.4587,
      "mean_token_accuracy": 0.16955531090497972,
      "num_tokens": 35210711.0,
      "step": 16965
    },
    {
      "entropy": 5.761061382293701,
      "epoch": 1.5393686502177069,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047668282210679825,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1633588433265686,
      "num_tokens": 35220866.0,
      "step": 16970
    },
    {
      "entropy": 5.750969362258911,
      "epoch": 1.5398222060957911,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004766684760795745,
      "loss": 5.4435,
      "mean_token_accuracy": 0.16699933111667634,
      "num_tokens": 35231478.0,
      "step": 16975
    },
    {
      "entropy": 5.692089033126831,
      "epoch": 1.5402757619738752,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047665412588179524,
      "loss": 5.3978,
      "mean_token_accuracy": 0.1731133669614792,
      "num_tokens": 35242233.0,
      "step": 16980
    },
    {
      "entropy": 5.672804975509644,
      "epoch": 1.5407293178519592,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004766397715137573,
      "loss": 5.4511,
      "mean_token_accuracy": 0.1641622394323349,
      "num_tokens": 35253097.0,
      "step": 16985
    },
    {
      "entropy": 5.829147672653198,
      "epoch": 1.5411828737300435,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047662541297575743,
      "loss": 5.5921,
      "mean_token_accuracy": 0.1655561864376068,
      "num_tokens": 35264258.0,
      "step": 16990
    },
    {
      "entropy": 5.878200483322144,
      "epoch": 1.5416364296081277,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047661105026809274,
      "loss": 5.5831,
      "mean_token_accuracy": 0.15921717882156372,
      "num_tokens": 35273501.0,
      "step": 16995
    },
    {
      "entropy": 5.732240295410156,
      "epoch": 1.542089985486212,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004765966833910601,
      "loss": 5.5231,
      "mean_token_accuracy": 0.16280568093061448,
      "num_tokens": 35284125.0,
      "step": 17000
    },
    {
      "entropy": 5.803600168228149,
      "epoch": 1.542543541364296,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004765823123449567,
      "loss": 5.5391,
      "mean_token_accuracy": 0.16041834354400636,
      "num_tokens": 35296039.0,
      "step": 17005
    },
    {
      "entropy": 5.821182203292847,
      "epoch": 1.54299709724238,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047656793713007965,
      "loss": 5.4938,
      "mean_token_accuracy": 0.15927931666374207,
      "num_tokens": 35306178.0,
      "step": 17010
    },
    {
      "entropy": 5.728965330123901,
      "epoch": 1.5434506531204644,
      "grad_norm": 3.984375,
      "learning_rate": 0.00047655355774672636,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16631783097982406,
      "num_tokens": 35316445.0,
      "step": 17015
    },
    {
      "entropy": 5.768176555633545,
      "epoch": 1.5439042089985486,
      "grad_norm": 7.75,
      "learning_rate": 0.00047653917419519413,
      "loss": 5.4803,
      "mean_token_accuracy": 0.16612288504838943,
      "num_tokens": 35327274.0,
      "step": 17020
    },
    {
      "entropy": 5.760219430923462,
      "epoch": 1.544357764876633,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004765247864757804,
      "loss": 5.4842,
      "mean_token_accuracy": 0.1660437136888504,
      "num_tokens": 35339007.0,
      "step": 17025
    },
    {
      "entropy": 5.806102609634399,
      "epoch": 1.544811320754717,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004765103945887827,
      "loss": 5.5917,
      "mean_token_accuracy": 0.1552738592028618,
      "num_tokens": 35349303.0,
      "step": 17030
    },
    {
      "entropy": 5.809821844100952,
      "epoch": 1.5452648766328012,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004764959985344987,
      "loss": 5.4375,
      "mean_token_accuracy": 0.16315661370754242,
      "num_tokens": 35359556.0,
      "step": 17035
    },
    {
      "entropy": 5.733943796157837,
      "epoch": 1.5457184325108853,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004764815983132261,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1724398538470268,
      "num_tokens": 35370043.0,
      "step": 17040
    },
    {
      "entropy": 5.722028493881226,
      "epoch": 1.5461719883889695,
      "grad_norm": 2.875,
      "learning_rate": 0.0004764671939252628,
      "loss": 5.4291,
      "mean_token_accuracy": 0.1702588304877281,
      "num_tokens": 35380031.0,
      "step": 17045
    },
    {
      "entropy": 5.712087488174438,
      "epoch": 1.5466255442670538,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004764527853709064,
      "loss": 5.5318,
      "mean_token_accuracy": 0.16298767924308777,
      "num_tokens": 35390481.0,
      "step": 17050
    },
    {
      "entropy": 5.735470867156982,
      "epoch": 1.5470791001451378,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004764383726504552,
      "loss": 5.5134,
      "mean_token_accuracy": 0.16611987054347993,
      "num_tokens": 35401756.0,
      "step": 17055
    },
    {
      "entropy": 5.766138076782227,
      "epoch": 1.547532656023222,
      "grad_norm": 7.9375,
      "learning_rate": 0.0004764239557642071,
      "loss": 5.4359,
      "mean_token_accuracy": 0.16979036331176758,
      "num_tokens": 35411045.0,
      "step": 17060
    },
    {
      "entropy": 5.747288846969605,
      "epoch": 1.5479862119013061,
      "grad_norm": 2.625,
      "learning_rate": 0.0004764095347124603,
      "loss": 5.4664,
      "mean_token_accuracy": 0.16777900159358977,
      "num_tokens": 35421256.0,
      "step": 17065
    },
    {
      "entropy": 5.769407272338867,
      "epoch": 1.5484397677793904,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004763951094955129,
      "loss": 5.4952,
      "mean_token_accuracy": 0.1699524700641632,
      "num_tokens": 35430798.0,
      "step": 17070
    },
    {
      "entropy": 5.731281042098999,
      "epoch": 1.5488933236574747,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004763806801136634,
      "loss": 5.4371,
      "mean_token_accuracy": 0.17484741955995559,
      "num_tokens": 35440459.0,
      "step": 17075
    },
    {
      "entropy": 5.77934160232544,
      "epoch": 1.549346879535559,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004763662465672101,
      "loss": 5.4847,
      "mean_token_accuracy": 0.16973735839128495,
      "num_tokens": 35451586.0,
      "step": 17080
    },
    {
      "entropy": 5.77488431930542,
      "epoch": 1.549800435413643,
      "grad_norm": 3.0,
      "learning_rate": 0.00047635180885645146,
      "loss": 5.5348,
      "mean_token_accuracy": 0.16148618310689927,
      "num_tokens": 35462197.0,
      "step": 17085
    },
    {
      "entropy": 5.749895906448364,
      "epoch": 1.550253991291727,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004763373669816862,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16848749071359634,
      "num_tokens": 35471737.0,
      "step": 17090
    },
    {
      "entropy": 5.78358645439148,
      "epoch": 1.5507075471698113,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004763229209432128,
      "loss": 5.5331,
      "mean_token_accuracy": 0.1620984598994255,
      "num_tokens": 35482016.0,
      "step": 17095
    },
    {
      "entropy": 5.790984106063843,
      "epoch": 1.5511611030478956,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004763084707413301,
      "loss": 5.607,
      "mean_token_accuracy": 0.1489139437675476,
      "num_tokens": 35492622.0,
      "step": 17100
    },
    {
      "entropy": 5.839964914321899,
      "epoch": 1.5516146589259798,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047629401637633696,
      "loss": 5.647,
      "mean_token_accuracy": 0.15655498951673508,
      "num_tokens": 35504468.0,
      "step": 17105
    },
    {
      "entropy": 5.803635740280152,
      "epoch": 1.5520682148040639,
      "grad_norm": 2.96875,
      "learning_rate": 0.00047627955784853226,
      "loss": 5.4367,
      "mean_token_accuracy": 0.16651920825242997,
      "num_tokens": 35514272.0,
      "step": 17110
    },
    {
      "entropy": 5.762120056152344,
      "epoch": 1.552521770682148,
      "grad_norm": 3.09375,
      "learning_rate": 0.000476265095158215,
      "loss": 5.4642,
      "mean_token_accuracy": 0.1617017701268196,
      "num_tokens": 35524110.0,
      "step": 17115
    },
    {
      "entropy": 5.726183891296387,
      "epoch": 1.5529753265602322,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004762506283056843,
      "loss": 5.4138,
      "mean_token_accuracy": 0.17430368959903716,
      "num_tokens": 35535504.0,
      "step": 17120
    },
    {
      "entropy": 5.859991455078125,
      "epoch": 1.5534288824383164,
      "grad_norm": 2.0,
      "learning_rate": 0.0004762361572912393,
      "loss": 5.5719,
      "mean_token_accuracy": 0.1573566287755966,
      "num_tokens": 35544885.0,
      "step": 17125
    },
    {
      "entropy": 5.79107403755188,
      "epoch": 1.5538824383164007,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047622168211517926,
      "loss": 5.5801,
      "mean_token_accuracy": 0.16384306252002717,
      "num_tokens": 35555485.0,
      "step": 17130
    },
    {
      "entropy": 5.711680173873901,
      "epoch": 1.5543359941944848,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047620720277780353,
      "loss": 5.411,
      "mean_token_accuracy": 0.1675860494375229,
      "num_tokens": 35567197.0,
      "step": 17135
    },
    {
      "entropy": 5.778192329406738,
      "epoch": 1.5547895500725688,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047619271927941165,
      "loss": 5.4581,
      "mean_token_accuracy": 0.16629468351602555,
      "num_tokens": 35577576.0,
      "step": 17140
    },
    {
      "entropy": 5.793490695953369,
      "epoch": 1.555243105950653,
      "grad_norm": 2.71875,
      "learning_rate": 0.000476178231620303,
      "loss": 5.5094,
      "mean_token_accuracy": 0.1627573013305664,
      "num_tokens": 35588012.0,
      "step": 17145
    },
    {
      "entropy": 5.737931108474731,
      "epoch": 1.5556966618287373,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004761637398007774,
      "loss": 5.4991,
      "mean_token_accuracy": 0.16023260951042176,
      "num_tokens": 35598076.0,
      "step": 17150
    },
    {
      "entropy": 5.648002481460571,
      "epoch": 1.5561502177068216,
      "grad_norm": 2.25,
      "learning_rate": 0.0004761492438211342,
      "loss": 5.3086,
      "mean_token_accuracy": 0.1730985701084137,
      "num_tokens": 35607170.0,
      "step": 17155
    },
    {
      "entropy": 5.740974569320679,
      "epoch": 1.5566037735849056,
      "grad_norm": 2.75,
      "learning_rate": 0.00047613474368167355,
      "loss": 5.4481,
      "mean_token_accuracy": 0.16214676946401596,
      "num_tokens": 35617865.0,
      "step": 17160
    },
    {
      "entropy": 5.780062675476074,
      "epoch": 1.5570573294629897,
      "grad_norm": 3.359375,
      "learning_rate": 0.000476120239382695,
      "loss": 5.5343,
      "mean_token_accuracy": 0.16312469393014908,
      "num_tokens": 35629329.0,
      "step": 17165
    },
    {
      "entropy": 5.728092432022095,
      "epoch": 1.557510885341074,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047610573092449877,
      "loss": 5.5004,
      "mean_token_accuracy": 0.1648014709353447,
      "num_tokens": 35640044.0,
      "step": 17170
    },
    {
      "entropy": 5.772489547729492,
      "epoch": 1.5579644412191582,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004760912183073847,
      "loss": 5.5204,
      "mean_token_accuracy": 0.16458816081285477,
      "num_tokens": 35650435.0,
      "step": 17175
    },
    {
      "entropy": 5.782632732391358,
      "epoch": 1.5584179970972425,
      "grad_norm": 2.703125,
      "learning_rate": 0.000476076701531653,
      "loss": 5.4252,
      "mean_token_accuracy": 0.16266798228025436,
      "num_tokens": 35659613.0,
      "step": 17180
    },
    {
      "entropy": 5.7409741401672365,
      "epoch": 1.5588715529753265,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004760621805976039,
      "loss": 5.4863,
      "mean_token_accuracy": 0.16694975644350052,
      "num_tokens": 35670987.0,
      "step": 17185
    },
    {
      "entropy": 5.742603492736817,
      "epoch": 1.5593251088534106,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047604765550553766,
      "loss": 5.4666,
      "mean_token_accuracy": 0.17111682891845703,
      "num_tokens": 35682505.0,
      "step": 17190
    },
    {
      "entropy": 5.76828236579895,
      "epoch": 1.5597786647314948,
      "grad_norm": 3.59375,
      "learning_rate": 0.00047603312625575466,
      "loss": 5.4447,
      "mean_token_accuracy": 0.17024013251066208,
      "num_tokens": 35693056.0,
      "step": 17195
    },
    {
      "entropy": 5.737306070327759,
      "epoch": 1.5602322206095791,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047601859284855544,
      "loss": 5.4254,
      "mean_token_accuracy": 0.1706910952925682,
      "num_tokens": 35703659.0,
      "step": 17200
    },
    {
      "entropy": 5.725793552398682,
      "epoch": 1.5606857764876634,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004760040552842403,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1613212361931801,
      "num_tokens": 35714367.0,
      "step": 17205
    },
    {
      "entropy": 5.69540114402771,
      "epoch": 1.5611393323657474,
      "grad_norm": 4.96875,
      "learning_rate": 0.00047598951356311035,
      "loss": 5.4157,
      "mean_token_accuracy": 0.1682739332318306,
      "num_tokens": 35724773.0,
      "step": 17210
    },
    {
      "entropy": 5.781873512268066,
      "epoch": 1.5615928882438317,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004759749676854658,
      "loss": 5.5863,
      "mean_token_accuracy": 0.16127568185329438,
      "num_tokens": 35735622.0,
      "step": 17215
    },
    {
      "entropy": 5.763715028762817,
      "epoch": 1.5620464441219157,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004759604176516079,
      "loss": 5.435,
      "mean_token_accuracy": 0.16790187507867813,
      "num_tokens": 35744967.0,
      "step": 17220
    },
    {
      "entropy": 5.763375616073608,
      "epoch": 1.5625,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004759458634618372,
      "loss": 5.5024,
      "mean_token_accuracy": 0.16670823395252227,
      "num_tokens": 35755647.0,
      "step": 17225
    },
    {
      "entropy": 5.77381477355957,
      "epoch": 1.5629535558780843,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004759313051164549,
      "loss": 5.5031,
      "mean_token_accuracy": 0.15969977974891664,
      "num_tokens": 35765595.0,
      "step": 17230
    },
    {
      "entropy": 5.808434200286865,
      "epoch": 1.5634071117561683,
      "grad_norm": 2.359375,
      "learning_rate": 0.000475916742615762,
      "loss": 5.4639,
      "mean_token_accuracy": 0.16989433616399766,
      "num_tokens": 35775804.0,
      "step": 17235
    },
    {
      "entropy": 5.735184955596924,
      "epoch": 1.5638606676342526,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004759021759600597,
      "loss": 5.4073,
      "mean_token_accuracy": 0.16915059685707093,
      "num_tokens": 35786227.0,
      "step": 17240
    },
    {
      "entropy": 5.787527322769165,
      "epoch": 1.5643142235123366,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004758876051496492,
      "loss": 5.5457,
      "mean_token_accuracy": 0.16405523419380189,
      "num_tokens": 35797511.0,
      "step": 17245
    },
    {
      "entropy": 5.803834104537964,
      "epoch": 1.5647677793904209,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047587303018483177,
      "loss": 5.4827,
      "mean_token_accuracy": 0.16339448392391204,
      "num_tokens": 35807248.0,
      "step": 17250
    },
    {
      "entropy": 5.73339467048645,
      "epoch": 1.5652213352685052,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047585845106590897,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16064847260713577,
      "num_tokens": 35818164.0,
      "step": 17255
    },
    {
      "entropy": 5.723453426361084,
      "epoch": 1.5656748911465894,
      "grad_norm": 3.34375,
      "learning_rate": 0.00047584386779318214,
      "loss": 5.4748,
      "mean_token_accuracy": 0.17431624084711075,
      "num_tokens": 35828746.0,
      "step": 17260
    },
    {
      "entropy": 5.86108169555664,
      "epoch": 1.5661284470246735,
      "grad_norm": 3.46875,
      "learning_rate": 0.000475829280366953,
      "loss": 5.5121,
      "mean_token_accuracy": 0.1605452388525009,
      "num_tokens": 35839325.0,
      "step": 17265
    },
    {
      "entropy": 5.766029262542725,
      "epoch": 1.5665820029027575,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047581468878752313,
      "loss": 5.451,
      "mean_token_accuracy": 0.16931989341974257,
      "num_tokens": 35849406.0,
      "step": 17270
    },
    {
      "entropy": 5.692113208770752,
      "epoch": 1.5670355587808418,
      "grad_norm": 3.484375,
      "learning_rate": 0.00047580009305519427,
      "loss": 5.3894,
      "mean_token_accuracy": 0.16529992520809172,
      "num_tokens": 35858715.0,
      "step": 17275
    },
    {
      "entropy": 5.696296691894531,
      "epoch": 1.567489114658926,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004757854931702684,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16766365766525268,
      "num_tokens": 35869063.0,
      "step": 17280
    },
    {
      "entropy": 5.700288534164429,
      "epoch": 1.5679426705370103,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047577088913304733,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1728409230709076,
      "num_tokens": 35878428.0,
      "step": 17285
    },
    {
      "entropy": 5.7382019519805905,
      "epoch": 1.5683962264150944,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047575628094383317,
      "loss": 5.4881,
      "mean_token_accuracy": 0.16865094602108002,
      "num_tokens": 35888801.0,
      "step": 17290
    },
    {
      "entropy": 5.691158390045166,
      "epoch": 1.5688497822931784,
      "grad_norm": 2.375,
      "learning_rate": 0.0004757416686029279,
      "loss": 5.4824,
      "mean_token_accuracy": 0.16762811094522476,
      "num_tokens": 35899381.0,
      "step": 17295
    },
    {
      "entropy": 5.816081953048706,
      "epoch": 1.5693033381712627,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004757270521106338,
      "loss": 5.529,
      "mean_token_accuracy": 0.16440595388412477,
      "num_tokens": 35909546.0,
      "step": 17300
    },
    {
      "entropy": 5.753160524368286,
      "epoch": 1.569756894049347,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004757124314672531,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1666400134563446,
      "num_tokens": 35919301.0,
      "step": 17305
    },
    {
      "entropy": 5.841774654388428,
      "epoch": 1.5702104499274312,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047569780667308816,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1635059431195259,
      "num_tokens": 35928603.0,
      "step": 17310
    },
    {
      "entropy": 5.768608283996582,
      "epoch": 1.5706640058055152,
      "grad_norm": 5.25,
      "learning_rate": 0.00047568317772844147,
      "loss": 5.482,
      "mean_token_accuracy": 0.169549460709095,
      "num_tokens": 35939005.0,
      "step": 17315
    },
    {
      "entropy": 5.782516431808472,
      "epoch": 1.5711175616835993,
      "grad_norm": 4.96875,
      "learning_rate": 0.00047566854463361544,
      "loss": 5.4754,
      "mean_token_accuracy": 0.17035190016031265,
      "num_tokens": 35948471.0,
      "step": 17320
    },
    {
      "entropy": 5.7585631847381595,
      "epoch": 1.5715711175616836,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004756539073889128,
      "loss": 5.511,
      "mean_token_accuracy": 0.1656458392739296,
      "num_tokens": 35958488.0,
      "step": 17325
    },
    {
      "entropy": 5.788969707489014,
      "epoch": 1.5720246734397678,
      "grad_norm": 5.1875,
      "learning_rate": 0.00047563926599463623,
      "loss": 5.5559,
      "mean_token_accuracy": 0.16097566336393357,
      "num_tokens": 35968291.0,
      "step": 17330
    },
    {
      "entropy": 5.789391946792603,
      "epoch": 1.572478229317852,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004756246204510886,
      "loss": 5.5054,
      "mean_token_accuracy": 0.1613545000553131,
      "num_tokens": 35978565.0,
      "step": 17335
    },
    {
      "entropy": 5.782929611206055,
      "epoch": 1.5729317851959361,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047560997075857253,
      "loss": 5.4225,
      "mean_token_accuracy": 0.17287506759166718,
      "num_tokens": 35987483.0,
      "step": 17340
    },
    {
      "entropy": 5.751060056686401,
      "epoch": 1.5733853410740202,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004755953169173913,
      "loss": 5.4821,
      "mean_token_accuracy": 0.16508522629737854,
      "num_tokens": 35998355.0,
      "step": 17345
    },
    {
      "entropy": 5.703908967971802,
      "epoch": 1.5738388969521044,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004755806589278477,
      "loss": 5.462,
      "mean_token_accuracy": 0.16875971406698226,
      "num_tokens": 36008365.0,
      "step": 17350
    },
    {
      "entropy": 5.7843098640441895,
      "epoch": 1.5742924528301887,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047556599679024497,
      "loss": 5.5406,
      "mean_token_accuracy": 0.15891460925340653,
      "num_tokens": 36018880.0,
      "step": 17355
    },
    {
      "entropy": 5.808669471740723,
      "epoch": 1.574746008708273,
      "grad_norm": 6.375,
      "learning_rate": 0.00047555133050488637,
      "loss": 5.5517,
      "mean_token_accuracy": 0.1589163661003113,
      "num_tokens": 36029335.0,
      "step": 17360
    },
    {
      "entropy": 5.722918844223022,
      "epoch": 1.575199564586357,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047553666007207516,
      "loss": 5.445,
      "mean_token_accuracy": 0.17202243953943253,
      "num_tokens": 36039774.0,
      "step": 17365
    },
    {
      "entropy": 5.8274835586547855,
      "epoch": 1.575653120464441,
      "grad_norm": 3.875,
      "learning_rate": 0.00047552198549211463,
      "loss": 5.5661,
      "mean_token_accuracy": 0.16887208968400955,
      "num_tokens": 36049864.0,
      "step": 17370
    },
    {
      "entropy": 5.799830913543701,
      "epoch": 1.5761066763425253,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004755073067653084,
      "loss": 5.484,
      "mean_token_accuracy": 0.1658618852496147,
      "num_tokens": 36059224.0,
      "step": 17375
    },
    {
      "entropy": 5.651365327835083,
      "epoch": 1.5765602322206096,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004754926238919599,
      "loss": 5.4244,
      "mean_token_accuracy": 0.1763281986117363,
      "num_tokens": 36069243.0,
      "step": 17380
    },
    {
      "entropy": 5.685806608200073,
      "epoch": 1.5770137880986939,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004754779368723729,
      "loss": 5.4676,
      "mean_token_accuracy": 0.1644148200750351,
      "num_tokens": 36079833.0,
      "step": 17385
    },
    {
      "entropy": 5.755171918869019,
      "epoch": 1.577467343976778,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004754632457068511,
      "loss": 5.4782,
      "mean_token_accuracy": 0.1648145541548729,
      "num_tokens": 36090673.0,
      "step": 17390
    },
    {
      "entropy": 5.774137926101685,
      "epoch": 1.5779208998548622,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004754485503956982,
      "loss": 5.5139,
      "mean_token_accuracy": 0.16774625927209855,
      "num_tokens": 36101719.0,
      "step": 17395
    },
    {
      "entropy": 5.751531553268433,
      "epoch": 1.5783744557329462,
      "grad_norm": 4.625,
      "learning_rate": 0.0004754338509392183,
      "loss": 5.4786,
      "mean_token_accuracy": 0.16290032640099525,
      "num_tokens": 36112352.0,
      "step": 17400
    },
    {
      "entropy": 5.656608581542969,
      "epoch": 1.5788280116110305,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004754191473377153,
      "loss": 5.3435,
      "mean_token_accuracy": 0.17451141923666,
      "num_tokens": 36121797.0,
      "step": 17405
    },
    {
      "entropy": 5.730385732650757,
      "epoch": 1.5792815674891147,
      "grad_norm": 2.875,
      "learning_rate": 0.00047540443959149316,
      "loss": 5.4866,
      "mean_token_accuracy": 0.16647754162549971,
      "num_tokens": 36132094.0,
      "step": 17410
    },
    {
      "entropy": 5.732181453704834,
      "epoch": 1.5797351233671988,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004753897277008562,
      "loss": 5.4697,
      "mean_token_accuracy": 0.1691714346408844,
      "num_tokens": 36142860.0,
      "step": 17415
    },
    {
      "entropy": 5.748935699462891,
      "epoch": 1.580188679245283,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004753750116661086,
      "loss": 5.4655,
      "mean_token_accuracy": 0.16277487426996232,
      "num_tokens": 36154555.0,
      "step": 17420
    },
    {
      "entropy": 5.818053102493286,
      "epoch": 1.580642235123367,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047536029148755477,
      "loss": 5.585,
      "mean_token_accuracy": 0.1570190370082855,
      "num_tokens": 36165479.0,
      "step": 17425
    },
    {
      "entropy": 5.788660144805908,
      "epoch": 1.5810957910014514,
      "grad_norm": 3.984375,
      "learning_rate": 0.00047534556716549893,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1626171126961708,
      "num_tokens": 36175873.0,
      "step": 17430
    },
    {
      "entropy": 5.771045732498169,
      "epoch": 1.5815493468795356,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047533083870024575,
      "loss": 5.4717,
      "mean_token_accuracy": 0.16632398664951326,
      "num_tokens": 36186792.0,
      "step": 17435
    },
    {
      "entropy": 5.862618112564087,
      "epoch": 1.58200290275762,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004753161060920998,
      "loss": 5.5837,
      "mean_token_accuracy": 0.1609506845474243,
      "num_tokens": 36196691.0,
      "step": 17440
    },
    {
      "entropy": 5.769778823852539,
      "epoch": 1.582456458635704,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004753013693413657,
      "loss": 5.4809,
      "mean_token_accuracy": 0.1651309221982956,
      "num_tokens": 36207173.0,
      "step": 17445
    },
    {
      "entropy": 5.667409038543701,
      "epoch": 1.582910014513788,
      "grad_norm": 4.40625,
      "learning_rate": 0.00047528662844834827,
      "loss": 5.479,
      "mean_token_accuracy": 0.16633691638708115,
      "num_tokens": 36217912.0,
      "step": 17450
    },
    {
      "entropy": 5.678972005844116,
      "epoch": 1.5833635703918723,
      "grad_norm": 3.25,
      "learning_rate": 0.00047527188341335233,
      "loss": 5.3885,
      "mean_token_accuracy": 0.17482518702745437,
      "num_tokens": 36227714.0,
      "step": 17455
    },
    {
      "entropy": 5.7722025394439695,
      "epoch": 1.5838171262699565,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004752571342366828,
      "loss": 5.4641,
      "mean_token_accuracy": 0.16554862707853318,
      "num_tokens": 36239114.0,
      "step": 17460
    },
    {
      "entropy": 5.726867818832398,
      "epoch": 1.5842706821480408,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004752423809186446,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1619683176279068,
      "num_tokens": 36249436.0,
      "step": 17465
    },
    {
      "entropy": 5.7600068092346195,
      "epoch": 1.5847242380261248,
      "grad_norm": 2.625,
      "learning_rate": 0.00047522762345954305,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16759982854127883,
      "num_tokens": 36259541.0,
      "step": 17470
    },
    {
      "entropy": 5.786817312240601,
      "epoch": 1.5851777939042089,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004752128618596832,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1604149341583252,
      "num_tokens": 36270689.0,
      "step": 17475
    },
    {
      "entropy": 5.829040813446045,
      "epoch": 1.5856313497822931,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047519809611937036,
      "loss": 5.5546,
      "mean_token_accuracy": 0.16309992969036102,
      "num_tokens": 36281191.0,
      "step": 17480
    },
    {
      "entropy": 5.657180738449097,
      "epoch": 1.5860849056603774,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004751833262389098,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1645319417119026,
      "num_tokens": 36291022.0,
      "step": 17485
    },
    {
      "entropy": 5.770510578155518,
      "epoch": 1.5865384615384617,
      "grad_norm": 3.296875,
      "learning_rate": 0.00047516855221860715,
      "loss": 5.5115,
      "mean_token_accuracy": 0.17102809399366378,
      "num_tokens": 36301237.0,
      "step": 17490
    },
    {
      "entropy": 5.835186338424682,
      "epoch": 1.5869920174165457,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047515377405876774,
      "loss": 5.4897,
      "mean_token_accuracy": 0.16277357190847397,
      "num_tokens": 36311792.0,
      "step": 17495
    },
    {
      "entropy": 5.692095994949341,
      "epoch": 1.5874455732946298,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004751389917596973,
      "loss": 5.4734,
      "mean_token_accuracy": 0.17079854458570481,
      "num_tokens": 36321847.0,
      "step": 17500
    },
    {
      "entropy": 5.697036266326904,
      "epoch": 1.587899129172714,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047512420532170154,
      "loss": 5.5418,
      "mean_token_accuracy": 0.16037049144506454,
      "num_tokens": 36332735.0,
      "step": 17505
    },
    {
      "entropy": 5.793141841888428,
      "epoch": 1.5883526850507983,
      "grad_norm": 4.125,
      "learning_rate": 0.0004751094147450862,
      "loss": 5.5621,
      "mean_token_accuracy": 0.16483381390571594,
      "num_tokens": 36343149.0,
      "step": 17510
    },
    {
      "entropy": 5.797153425216675,
      "epoch": 1.5888062409288826,
      "grad_norm": 2.875,
      "learning_rate": 0.0004750946200301571,
      "loss": 5.4472,
      "mean_token_accuracy": 0.17080761790275573,
      "num_tokens": 36353081.0,
      "step": 17515
    },
    {
      "entropy": 5.758509492874145,
      "epoch": 1.5892597968069666,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047507982117722035,
      "loss": 5.4977,
      "mean_token_accuracy": 0.1612115040421486,
      "num_tokens": 36363624.0,
      "step": 17520
    },
    {
      "entropy": 5.812560415267944,
      "epoch": 1.5897133526850507,
      "grad_norm": 2.484375,
      "learning_rate": 0.000475065018186582,
      "loss": 5.5189,
      "mean_token_accuracy": 0.16480037420988083,
      "num_tokens": 36374081.0,
      "step": 17525
    },
    {
      "entropy": 5.769211149215698,
      "epoch": 1.590166908563135,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047505021105854794,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1620366871356964,
      "num_tokens": 36384360.0,
      "step": 17530
    },
    {
      "entropy": 5.757202243804931,
      "epoch": 1.5906204644412192,
      "grad_norm": 7.6875,
      "learning_rate": 0.00047503539979342456,
      "loss": 5.5428,
      "mean_token_accuracy": 0.16642032861709594,
      "num_tokens": 36395924.0,
      "step": 17535
    },
    {
      "entropy": 5.679582166671753,
      "epoch": 1.5910740203193035,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004750205843915182,
      "loss": 5.3827,
      "mean_token_accuracy": 0.17185181826353074,
      "num_tokens": 36405246.0,
      "step": 17540
    },
    {
      "entropy": 5.771241044998169,
      "epoch": 1.5915275761973875,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004750057648531351,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16783628314733506,
      "num_tokens": 36414964.0,
      "step": 17545
    },
    {
      "entropy": 5.767411422729492,
      "epoch": 1.5919811320754715,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004749909411785819,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1650505244731903,
      "num_tokens": 36425011.0,
      "step": 17550
    },
    {
      "entropy": 5.6866672992706295,
      "epoch": 1.5924346879535558,
      "grad_norm": 3.34375,
      "learning_rate": 0.000474976113368165,
      "loss": 5.3983,
      "mean_token_accuracy": 0.17585941255092621,
      "num_tokens": 36434734.0,
      "step": 17555
    },
    {
      "entropy": 5.787780046463013,
      "epoch": 1.59288824383164,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004749612814221912,
      "loss": 5.4891,
      "mean_token_accuracy": 0.16718265861272813,
      "num_tokens": 36444748.0,
      "step": 17560
    },
    {
      "entropy": 5.840566349029541,
      "epoch": 1.5933417997097243,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047494644534096703,
      "loss": 5.5151,
      "mean_token_accuracy": 0.16032207757234573,
      "num_tokens": 36455076.0,
      "step": 17565
    },
    {
      "entropy": 5.743116569519043,
      "epoch": 1.5937953555878084,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004749316051247995,
      "loss": 5.4279,
      "mean_token_accuracy": 0.17147312462329864,
      "num_tokens": 36465302.0,
      "step": 17570
    },
    {
      "entropy": 5.759064626693726,
      "epoch": 1.5942489114658926,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047491676077399543,
      "loss": 5.5446,
      "mean_token_accuracy": 0.1631319046020508,
      "num_tokens": 36476666.0,
      "step": 17575
    },
    {
      "entropy": 5.72057147026062,
      "epoch": 1.5947024673439767,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004749019122888618,
      "loss": 5.4123,
      "mean_token_accuracy": 0.16968216449022294,
      "num_tokens": 36487379.0,
      "step": 17580
    },
    {
      "entropy": 5.710862922668457,
      "epoch": 1.595156023222061,
      "grad_norm": 4.75,
      "learning_rate": 0.00047488705966970565,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1748661607503891,
      "num_tokens": 36497897.0,
      "step": 17585
    },
    {
      "entropy": 5.762555265426636,
      "epoch": 1.5956095791001452,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004748722029168342,
      "loss": 5.5467,
      "mean_token_accuracy": 0.17152152508497237,
      "num_tokens": 36508528.0,
      "step": 17590
    },
    {
      "entropy": 5.822070550918579,
      "epoch": 1.5960631349782293,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004748573420305546,
      "loss": 5.524,
      "mean_token_accuracy": 0.16402218341827393,
      "num_tokens": 36519196.0,
      "step": 17595
    },
    {
      "entropy": 5.734179973602295,
      "epoch": 1.5965166908563135,
      "grad_norm": 5.71875,
      "learning_rate": 0.00047484247701117427,
      "loss": 5.5239,
      "mean_token_accuracy": 0.16501554548740388,
      "num_tokens": 36529871.0,
      "step": 17600
    },
    {
      "entropy": 5.854640817642212,
      "epoch": 1.5969702467343976,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004748276078590006,
      "loss": 5.5647,
      "mean_token_accuracy": 0.15793327391147613,
      "num_tokens": 36541188.0,
      "step": 17605
    },
    {
      "entropy": 5.8760590076446535,
      "epoch": 1.5974238026124818,
      "grad_norm": 3.125,
      "learning_rate": 0.0004748127345743411,
      "loss": 5.5657,
      "mean_token_accuracy": 0.16430289447307586,
      "num_tokens": 36551968.0,
      "step": 17610
    },
    {
      "entropy": 5.805030679702758,
      "epoch": 1.5978773584905661,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004747978571575033,
      "loss": 5.5046,
      "mean_token_accuracy": 0.16406284272670746,
      "num_tokens": 36562870.0,
      "step": 17615
    },
    {
      "entropy": 5.705535650253296,
      "epoch": 1.5983309143686504,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004747829756087949,
      "loss": 5.4371,
      "mean_token_accuracy": 0.17070819288492203,
      "num_tokens": 36573295.0,
      "step": 17620
    },
    {
      "entropy": 5.683192253112793,
      "epoch": 1.5987844702467344,
      "grad_norm": 2.75,
      "learning_rate": 0.00047476808992852364,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17305488139390945,
      "num_tokens": 36583830.0,
      "step": 17625
    },
    {
      "entropy": 5.7556885242462155,
      "epoch": 1.5992380261248185,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004747532001169974,
      "loss": 5.4718,
      "mean_token_accuracy": 0.17121025323867797,
      "num_tokens": 36594052.0,
      "step": 17630
    },
    {
      "entropy": 5.805377101898193,
      "epoch": 1.5996915820029027,
      "grad_norm": 16.75,
      "learning_rate": 0.00047473830617452415,
      "loss": 5.517,
      "mean_token_accuracy": 0.16405150443315505,
      "num_tokens": 36604378.0,
      "step": 17635
    },
    {
      "entropy": 5.811084318161011,
      "epoch": 1.600145137880987,
      "grad_norm": 3.96875,
      "learning_rate": 0.00047472340810141163,
      "loss": 5.608,
      "mean_token_accuracy": 0.15336571559309958,
      "num_tokens": 36615588.0,
      "step": 17640
    },
    {
      "entropy": 5.828241395950317,
      "epoch": 1.6005986937590713,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047470850589796833,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1609049141407013,
      "num_tokens": 36626369.0,
      "step": 17645
    },
    {
      "entropy": 5.81109504699707,
      "epoch": 1.6010522496371553,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047469359956450215,
      "loss": 5.5351,
      "mean_token_accuracy": 0.1608123540878296,
      "num_tokens": 36636565.0,
      "step": 17650
    },
    {
      "entropy": 5.820064544677734,
      "epoch": 1.6015058055152394,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047467868910132144,
      "loss": 5.5872,
      "mean_token_accuracy": 0.15833139419555664,
      "num_tokens": 36647721.0,
      "step": 17655
    },
    {
      "entropy": 5.744867897033691,
      "epoch": 1.6019593613933236,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004746637745087346,
      "loss": 5.3965,
      "mean_token_accuracy": 0.17056546211242676,
      "num_tokens": 36657887.0,
      "step": 17660
    },
    {
      "entropy": 5.804804086685181,
      "epoch": 1.6024129172714079,
      "grad_norm": 6.3125,
      "learning_rate": 0.00047464885578705,
      "loss": 5.4886,
      "mean_token_accuracy": 0.16347549855709076,
      "num_tokens": 36667918.0,
      "step": 17665
    },
    {
      "entropy": 5.665347623825073,
      "epoch": 1.6028664731494922,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047463393293657616,
      "loss": 5.4422,
      "mean_token_accuracy": 0.17264218777418136,
      "num_tokens": 36678491.0,
      "step": 17670
    },
    {
      "entropy": 5.721580362319946,
      "epoch": 1.6033200290275762,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004746190059576217,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1648408904671669,
      "num_tokens": 36688638.0,
      "step": 17675
    },
    {
      "entropy": 5.822503995895386,
      "epoch": 1.6037735849056602,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004746040748504954,
      "loss": 5.5151,
      "mean_token_accuracy": 0.16779906898736954,
      "num_tokens": 36699344.0,
      "step": 17680
    },
    {
      "entropy": 5.750469636917114,
      "epoch": 1.6042271407837445,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047458913961550593,
      "loss": 5.444,
      "mean_token_accuracy": 0.16815478056669236,
      "num_tokens": 36709606.0,
      "step": 17685
    },
    {
      "entropy": 5.677337503433227,
      "epoch": 1.6046806966618288,
      "grad_norm": 6.6875,
      "learning_rate": 0.0004745742002529622,
      "loss": 5.4104,
      "mean_token_accuracy": 0.16759033650159835,
      "num_tokens": 36718740.0,
      "step": 17690
    },
    {
      "entropy": 5.738113975524902,
      "epoch": 1.605134252539913,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004745592567631731,
      "loss": 5.4946,
      "mean_token_accuracy": 0.16052908301353455,
      "num_tokens": 36729960.0,
      "step": 17695
    },
    {
      "entropy": 5.644214200973511,
      "epoch": 1.605587808417997,
      "grad_norm": 2.5,
      "learning_rate": 0.00047454430914644773,
      "loss": 5.4076,
      "mean_token_accuracy": 0.17032331675291063,
      "num_tokens": 36740221.0,
      "step": 17700
    },
    {
      "entropy": 5.67592134475708,
      "epoch": 1.6060413642960811,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004745293574030952,
      "loss": 5.4307,
      "mean_token_accuracy": 0.17126678824424743,
      "num_tokens": 36750841.0,
      "step": 17705
    },
    {
      "entropy": 5.7318285465240475,
      "epoch": 1.6064949201741654,
      "grad_norm": 3.765625,
      "learning_rate": 0.00047451440153342477,
      "loss": 5.4411,
      "mean_token_accuracy": 0.1633999913930893,
      "num_tokens": 36760944.0,
      "step": 17710
    },
    {
      "entropy": 5.793593788146973,
      "epoch": 1.6069484760522497,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004744994415377457,
      "loss": 5.467,
      "mean_token_accuracy": 0.16669370085000992,
      "num_tokens": 36770922.0,
      "step": 17715
    },
    {
      "entropy": 5.762706661224366,
      "epoch": 1.607402031930334,
      "grad_norm": 3.921875,
      "learning_rate": 0.00047448447741636724,
      "loss": 5.4859,
      "mean_token_accuracy": 0.16215070635080336,
      "num_tokens": 36780702.0,
      "step": 17720
    },
    {
      "entropy": 5.759842300415039,
      "epoch": 1.607855587808418,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047446950916959905,
      "loss": 5.5296,
      "mean_token_accuracy": 0.15990192592144012,
      "num_tokens": 36791868.0,
      "step": 17725
    },
    {
      "entropy": 5.792073917388916,
      "epoch": 1.608309143686502,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004744545367977505,
      "loss": 5.5354,
      "mean_token_accuracy": 0.16470289081335068,
      "num_tokens": 36802330.0,
      "step": 17730
    },
    {
      "entropy": 5.811792087554932,
      "epoch": 1.6087626995645863,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047443956030113134,
      "loss": 5.5424,
      "mean_token_accuracy": 0.15985184162855148,
      "num_tokens": 36812936.0,
      "step": 17735
    },
    {
      "entropy": 5.779332828521729,
      "epoch": 1.6092162554426706,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047442457968005123,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1660194516181946,
      "num_tokens": 36823123.0,
      "step": 17740
    },
    {
      "entropy": 5.748415088653564,
      "epoch": 1.6096698113207548,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047440959493481996,
      "loss": 5.3985,
      "mean_token_accuracy": 0.17351642847061158,
      "num_tokens": 36833050.0,
      "step": 17745
    },
    {
      "entropy": 5.729624032974243,
      "epoch": 1.6101233671988389,
      "grad_norm": 2.125,
      "learning_rate": 0.00047439460606574745,
      "loss": 5.5344,
      "mean_token_accuracy": 0.16772380769252776,
      "num_tokens": 36842520.0,
      "step": 17750
    },
    {
      "entropy": 5.681790113449097,
      "epoch": 1.6105769230769231,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004743796130731437,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16364991664886475,
      "num_tokens": 36853024.0,
      "step": 17755
    },
    {
      "entropy": 5.751934003829956,
      "epoch": 1.6110304789550072,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004743646159573188,
      "loss": 5.4352,
      "mean_token_accuracy": 0.17198805660009384,
      "num_tokens": 36862534.0,
      "step": 17760
    },
    {
      "entropy": 5.772487640380859,
      "epoch": 1.6114840348330914,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004743496147185828,
      "loss": 5.425,
      "mean_token_accuracy": 0.1689123660326004,
      "num_tokens": 36872356.0,
      "step": 17765
    },
    {
      "entropy": 5.6932690143585205,
      "epoch": 1.6119375907111757,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047433460935724595,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1674752116203308,
      "num_tokens": 36882643.0,
      "step": 17770
    },
    {
      "entropy": 5.727265644073486,
      "epoch": 1.6123911465892597,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004743195998736186,
      "loss": 5.4847,
      "mean_token_accuracy": 0.16293875575065614,
      "num_tokens": 36893349.0,
      "step": 17775
    },
    {
      "entropy": 5.854079008102417,
      "epoch": 1.612844702467344,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047430458626801107,
      "loss": 5.4879,
      "mean_token_accuracy": 0.1652977094054222,
      "num_tokens": 36903713.0,
      "step": 17780
    },
    {
      "entropy": 5.752238130569458,
      "epoch": 1.613298258345428,
      "grad_norm": 2.625,
      "learning_rate": 0.000474289568540734,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1696855068206787,
      "num_tokens": 36912999.0,
      "step": 17785
    },
    {
      "entropy": 5.770322036743164,
      "epoch": 1.6137518142235123,
      "grad_norm": 2.875,
      "learning_rate": 0.00047427454669209785,
      "loss": 5.5326,
      "mean_token_accuracy": 0.1647559642791748,
      "num_tokens": 36923598.0,
      "step": 17790
    },
    {
      "entropy": 5.770463275909424,
      "epoch": 1.6142053701015966,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047425952072241326,
      "loss": 5.4161,
      "mean_token_accuracy": 0.1639949917793274,
      "num_tokens": 36933036.0,
      "step": 17795
    },
    {
      "entropy": 5.771493816375733,
      "epoch": 1.6146589259796809,
      "grad_norm": 2.796875,
      "learning_rate": 0.000474244490631991,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16591634303331376,
      "num_tokens": 36943137.0,
      "step": 17800
    },
    {
      "entropy": 5.825366401672364,
      "epoch": 1.615112481857765,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004742294564211419,
      "loss": 5.6664,
      "mean_token_accuracy": 0.1587715819478035,
      "num_tokens": 36955619.0,
      "step": 17805
    },
    {
      "entropy": 5.779758930206299,
      "epoch": 1.615566037735849,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004742144180901769,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1734271302819252,
      "num_tokens": 36966377.0,
      "step": 17810
    },
    {
      "entropy": 5.851308059692383,
      "epoch": 1.6160195936139332,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047419937563940697,
      "loss": 5.4884,
      "mean_token_accuracy": 0.16991155296564103,
      "num_tokens": 36977100.0,
      "step": 17815
    },
    {
      "entropy": 5.764162969589234,
      "epoch": 1.6164731494920175,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047418432906914314,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1648203343153,
      "num_tokens": 36987120.0,
      "step": 17820
    },
    {
      "entropy": 5.767458248138428,
      "epoch": 1.6169267053701017,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004741692783796967,
      "loss": 5.5156,
      "mean_token_accuracy": 0.17140409499406814,
      "num_tokens": 36996885.0,
      "step": 17825
    },
    {
      "entropy": 5.742390775680542,
      "epoch": 1.6173802612481858,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047415422357137876,
      "loss": 5.4491,
      "mean_token_accuracy": 0.17157105207443238,
      "num_tokens": 37006570.0,
      "step": 17830
    },
    {
      "entropy": 5.83612813949585,
      "epoch": 1.6178338171262698,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004741391646445008,
      "loss": 5.533,
      "mean_token_accuracy": 0.16308058351278304,
      "num_tokens": 37017185.0,
      "step": 17835
    },
    {
      "entropy": 5.770243835449219,
      "epoch": 1.618287373004354,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004741241015993741,
      "loss": 5.5176,
      "mean_token_accuracy": 0.16109621822834014,
      "num_tokens": 37028386.0,
      "step": 17840
    },
    {
      "entropy": 5.713708877563477,
      "epoch": 1.6187409288824384,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047410903443631026,
      "loss": 5.3972,
      "mean_token_accuracy": 0.16598420739173889,
      "num_tokens": 37038326.0,
      "step": 17845
    },
    {
      "entropy": 5.810881042480469,
      "epoch": 1.6191944847605226,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004740939631556209,
      "loss": 5.513,
      "mean_token_accuracy": 0.16800838261842727,
      "num_tokens": 37048570.0,
      "step": 17850
    },
    {
      "entropy": 5.710151338577271,
      "epoch": 1.6196480406386067,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004740788877576176,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1745910093188286,
      "num_tokens": 37058909.0,
      "step": 17855
    },
    {
      "entropy": 5.771472311019897,
      "epoch": 1.6201015965166907,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004740638082426121,
      "loss": 5.462,
      "mean_token_accuracy": 0.17222315222024917,
      "num_tokens": 37068957.0,
      "step": 17860
    },
    {
      "entropy": 5.788452196121216,
      "epoch": 1.620555152394775,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047404872461091643,
      "loss": 5.4362,
      "mean_token_accuracy": 0.1635264202952385,
      "num_tokens": 37080187.0,
      "step": 17865
    },
    {
      "entropy": 5.739633178710937,
      "epoch": 1.6210087082728593,
      "grad_norm": 8.5625,
      "learning_rate": 0.00047403363686284236,
      "loss": 5.5059,
      "mean_token_accuracy": 0.16230847239494323,
      "num_tokens": 37090314.0,
      "step": 17870
    },
    {
      "entropy": 5.776074981689453,
      "epoch": 1.6214622641509435,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047401854499870197,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1652637541294098,
      "num_tokens": 37099907.0,
      "step": 17875
    },
    {
      "entropy": 5.749589490890503,
      "epoch": 1.6219158200290276,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047400344901880735,
      "loss": 5.5376,
      "mean_token_accuracy": 0.15882395207881927,
      "num_tokens": 37110495.0,
      "step": 17880
    },
    {
      "entropy": 5.830996370315551,
      "epoch": 1.6223693759071116,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004739883489234707,
      "loss": 5.5355,
      "mean_token_accuracy": 0.16459348499774934,
      "num_tokens": 37121957.0,
      "step": 17885
    },
    {
      "entropy": 5.803467226028443,
      "epoch": 1.6228229317851959,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047397324471300424,
      "loss": 5.4453,
      "mean_token_accuracy": 0.16810328364372254,
      "num_tokens": 37132743.0,
      "step": 17890
    },
    {
      "entropy": 5.779066181182861,
      "epoch": 1.6232764876632801,
      "grad_norm": 2.25,
      "learning_rate": 0.00047395813638772047,
      "loss": 5.5537,
      "mean_token_accuracy": 0.16044283658266068,
      "num_tokens": 37143947.0,
      "step": 17895
    },
    {
      "entropy": 5.867018747329712,
      "epoch": 1.6237300435413644,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047394302394793164,
      "loss": 5.6172,
      "mean_token_accuracy": 0.15855293571949006,
      "num_tokens": 37154316.0,
      "step": 17900
    },
    {
      "entropy": 5.769827175140381,
      "epoch": 1.6241835994194485,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004739279073939504,
      "loss": 5.4915,
      "mean_token_accuracy": 0.16654011309146882,
      "num_tokens": 37164999.0,
      "step": 17905
    },
    {
      "entropy": 5.81049861907959,
      "epoch": 1.6246371552975325,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004739127867260894,
      "loss": 5.5397,
      "mean_token_accuracy": 0.16377749145030976,
      "num_tokens": 37174392.0,
      "step": 17910
    },
    {
      "entropy": 5.729914236068725,
      "epoch": 1.6250907111756168,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004738976619446613,
      "loss": 5.5202,
      "mean_token_accuracy": 0.1657993823289871,
      "num_tokens": 37185341.0,
      "step": 17915
    },
    {
      "entropy": 5.83682107925415,
      "epoch": 1.625544267053701,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047388253304997874,
      "loss": 5.5087,
      "mean_token_accuracy": 0.15921589881181716,
      "num_tokens": 37196725.0,
      "step": 17920
    },
    {
      "entropy": 5.751535701751709,
      "epoch": 1.6259978229317853,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004738674000423548,
      "loss": 5.41,
      "mean_token_accuracy": 0.17050351053476334,
      "num_tokens": 37206877.0,
      "step": 17925
    },
    {
      "entropy": 5.656973028182984,
      "epoch": 1.6264513788098693,
      "grad_norm": 2.375,
      "learning_rate": 0.0004738522629221023,
      "loss": 5.3876,
      "mean_token_accuracy": 0.17137402892112732,
      "num_tokens": 37217277.0,
      "step": 17930
    },
    {
      "entropy": 5.69054217338562,
      "epoch": 1.6269049346879536,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004738371216895344,
      "loss": 5.4084,
      "mean_token_accuracy": 0.1685428127646446,
      "num_tokens": 37227716.0,
      "step": 17935
    },
    {
      "entropy": 5.717661190032959,
      "epoch": 1.6273584905660377,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004738219763449641,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1624217376112938,
      "num_tokens": 37238178.0,
      "step": 17940
    },
    {
      "entropy": 5.78687481880188,
      "epoch": 1.627812046444122,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004738068268887047,
      "loss": 5.505,
      "mean_token_accuracy": 0.1709125205874443,
      "num_tokens": 37248408.0,
      "step": 17945
    },
    {
      "entropy": 5.76186203956604,
      "epoch": 1.6282656023222062,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004737916733210694,
      "loss": 5.5926,
      "mean_token_accuracy": 0.1628739655017853,
      "num_tokens": 37259176.0,
      "step": 17950
    },
    {
      "entropy": 5.761851167678833,
      "epoch": 1.6287191582002902,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047377651564237165,
      "loss": 5.3924,
      "mean_token_accuracy": 0.17642752975225448,
      "num_tokens": 37269205.0,
      "step": 17955
    },
    {
      "entropy": 5.765105247497559,
      "epoch": 1.6291727140783745,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047376135385292494,
      "loss": 5.5236,
      "mean_token_accuracy": 0.17332034409046174,
      "num_tokens": 37278841.0,
      "step": 17960
    },
    {
      "entropy": 5.7521566390991214,
      "epoch": 1.6296262699564585,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004737461879530428,
      "loss": 5.5443,
      "mean_token_accuracy": 0.16545841544866563,
      "num_tokens": 37288487.0,
      "step": 17965
    },
    {
      "entropy": 5.798821926116943,
      "epoch": 1.6300798258345428,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047373101794303884,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16519451290369033,
      "num_tokens": 37300086.0,
      "step": 17970
    },
    {
      "entropy": 5.787726497650146,
      "epoch": 1.630533381712627,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004737158438232267,
      "loss": 5.5364,
      "mean_token_accuracy": 0.15512679666280746,
      "num_tokens": 37310072.0,
      "step": 17975
    },
    {
      "entropy": 5.804882192611695,
      "epoch": 1.6309869375907113,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047370066559392033,
      "loss": 5.5015,
      "mean_token_accuracy": 0.17356933951377868,
      "num_tokens": 37320278.0,
      "step": 17980
    },
    {
      "entropy": 5.722419595718383,
      "epoch": 1.6314404934687954,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047368548325543366,
      "loss": 5.4282,
      "mean_token_accuracy": 0.16354026943445205,
      "num_tokens": 37330741.0,
      "step": 17985
    },
    {
      "entropy": 5.733390855789184,
      "epoch": 1.6318940493468794,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004736702968080804,
      "loss": 5.514,
      "mean_token_accuracy": 0.16552259624004365,
      "num_tokens": 37341213.0,
      "step": 17990
    },
    {
      "entropy": 5.714855289459228,
      "epoch": 1.6323476052249637,
      "grad_norm": 3.546875,
      "learning_rate": 0.00047365510625217483,
      "loss": 5.4002,
      "mean_token_accuracy": 0.17041305005550383,
      "num_tokens": 37350528.0,
      "step": 17995
    },
    {
      "entropy": 5.753640031814575,
      "epoch": 1.632801161103048,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047363991158803117,
      "loss": 5.4823,
      "mean_token_accuracy": 0.15914918333292008,
      "num_tokens": 37359938.0,
      "step": 18000
    },
    {
      "epoch": 1.632801161103048,
      "eval_entropy": 5.550096890289859,
      "eval_loss": 5.53005313873291,
      "eval_mean_token_accuracy": 0.17172270843093346,
      "eval_num_tokens": 37359938.0,
      "eval_runtime": 26.3944,
      "eval_samples_per_second": 1339.679,
      "eval_steps_per_second": 167.46,
      "step": 18000
    },
    {
      "entropy": 5.754616069793701,
      "epoch": 1.6332547169811322,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047362471281596334,
      "loss": 5.4589,
      "mean_token_accuracy": 0.16467181891202926,
      "num_tokens": 37369744.0,
      "step": 18005
    },
    {
      "entropy": 5.808965063095092,
      "epoch": 1.6337082728592163,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047360950993628585,
      "loss": 5.5764,
      "mean_token_accuracy": 0.16452402770519256,
      "num_tokens": 37380578.0,
      "step": 18010
    },
    {
      "entropy": 5.746728849411011,
      "epoch": 1.6341618287373003,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004735943029493132,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16235966086387635,
      "num_tokens": 37390237.0,
      "step": 18015
    },
    {
      "entropy": 5.759697532653808,
      "epoch": 1.6346153846153846,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047357909185535965,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16467203199863434,
      "num_tokens": 37400261.0,
      "step": 18020
    },
    {
      "entropy": 5.755236434936523,
      "epoch": 1.6350689404934688,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004735638766547399,
      "loss": 5.4857,
      "mean_token_accuracy": 0.16921688616275787,
      "num_tokens": 37410012.0,
      "step": 18025
    },
    {
      "entropy": 5.79714183807373,
      "epoch": 1.6355224963715531,
      "grad_norm": 2.625,
      "learning_rate": 0.00047354865734776857,
      "loss": 5.5875,
      "mean_token_accuracy": 0.15842805355787276,
      "num_tokens": 37420530.0,
      "step": 18030
    },
    {
      "entropy": 5.811334085464478,
      "epoch": 1.6359760522496372,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004735334339347605,
      "loss": 5.4557,
      "mean_token_accuracy": 0.175724820792675,
      "num_tokens": 37431146.0,
      "step": 18035
    },
    {
      "entropy": 5.791562700271607,
      "epoch": 1.6364296081277212,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047351820641603035,
      "loss": 5.4223,
      "mean_token_accuracy": 0.1727816417813301,
      "num_tokens": 37441253.0,
      "step": 18040
    },
    {
      "entropy": 5.648637485504151,
      "epoch": 1.6368831640058055,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004735029747918931,
      "loss": 5.4591,
      "mean_token_accuracy": 0.16412138491868972,
      "num_tokens": 37451326.0,
      "step": 18045
    },
    {
      "entropy": 5.690817213058471,
      "epoch": 1.6373367198838897,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047348773906266374,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1709284797310829,
      "num_tokens": 37462462.0,
      "step": 18050
    },
    {
      "entropy": 5.7136290073394775,
      "epoch": 1.637790275761974,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004734724992286574,
      "loss": 5.4283,
      "mean_token_accuracy": 0.16969010531902312,
      "num_tokens": 37473016.0,
      "step": 18055
    },
    {
      "entropy": 5.792579507827758,
      "epoch": 1.638243831640058,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004734572552901892,
      "loss": 5.5523,
      "mean_token_accuracy": 0.16435847878456117,
      "num_tokens": 37483737.0,
      "step": 18060
    },
    {
      "entropy": 5.706974744796753,
      "epoch": 1.638697387518142,
      "grad_norm": 6.625,
      "learning_rate": 0.0004734420072475743,
      "loss": 5.4407,
      "mean_token_accuracy": 0.1673683002591133,
      "num_tokens": 37493070.0,
      "step": 18065
    },
    {
      "entropy": 5.729290437698364,
      "epoch": 1.6391509433962264,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004734267551011282,
      "loss": 5.4721,
      "mean_token_accuracy": 0.16383899450302125,
      "num_tokens": 37503696.0,
      "step": 18070
    },
    {
      "entropy": 5.685934734344483,
      "epoch": 1.6396044992743106,
      "grad_norm": 5.75,
      "learning_rate": 0.0004734114988511662,
      "loss": 5.4659,
      "mean_token_accuracy": 0.16844111382961274,
      "num_tokens": 37513468.0,
      "step": 18075
    },
    {
      "entropy": 5.6940430164337155,
      "epoch": 1.640058055152395,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004733962384980039,
      "loss": 5.3441,
      "mean_token_accuracy": 0.17803383022546768,
      "num_tokens": 37523631.0,
      "step": 18080
    },
    {
      "entropy": 5.817456197738648,
      "epoch": 1.640511611030479,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004733809740419568,
      "loss": 5.52,
      "mean_token_accuracy": 0.16899262368679047,
      "num_tokens": 37534602.0,
      "step": 18085
    },
    {
      "entropy": 5.741984081268311,
      "epoch": 1.640965166908563,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047336570548334053,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1659794494509697,
      "num_tokens": 37544547.0,
      "step": 18090
    },
    {
      "entropy": 5.748867082595825,
      "epoch": 1.6414187227866472,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047335043282247105,
      "loss": 5.4713,
      "mean_token_accuracy": 0.16318271011114122,
      "num_tokens": 37554903.0,
      "step": 18095
    },
    {
      "entropy": 5.727425813674927,
      "epoch": 1.6418722786647315,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047333515605966396,
      "loss": 5.3923,
      "mean_token_accuracy": 0.16589023917913437,
      "num_tokens": 37565586.0,
      "step": 18100
    },
    {
      "entropy": 5.781137466430664,
      "epoch": 1.6423258345428158,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047331987519523533,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16953398585319518,
      "num_tokens": 37575996.0,
      "step": 18105
    },
    {
      "entropy": 5.713779163360596,
      "epoch": 1.6427793904208998,
      "grad_norm": 4.375,
      "learning_rate": 0.0004733045902295012,
      "loss": 5.3802,
      "mean_token_accuracy": 0.1663781762123108,
      "num_tokens": 37586441.0,
      "step": 18110
    },
    {
      "entropy": 5.767160511016845,
      "epoch": 1.643232946298984,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047328930116277755,
      "loss": 5.5117,
      "mean_token_accuracy": 0.1651667356491089,
      "num_tokens": 37598226.0,
      "step": 18115
    },
    {
      "entropy": 5.789662408828735,
      "epoch": 1.6436865021770681,
      "grad_norm": 4.125,
      "learning_rate": 0.0004732740079953806,
      "loss": 5.4981,
      "mean_token_accuracy": 0.16533599346876143,
      "num_tokens": 37609102.0,
      "step": 18120
    },
    {
      "entropy": 5.722034931182861,
      "epoch": 1.6441400580551524,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004732587107276267,
      "loss": 5.4838,
      "mean_token_accuracy": 0.16687125265598296,
      "num_tokens": 37619349.0,
      "step": 18125
    },
    {
      "entropy": 5.714503812789917,
      "epoch": 1.6445936139332367,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047324340935983225,
      "loss": 5.477,
      "mean_token_accuracy": 0.1727244585752487,
      "num_tokens": 37629844.0,
      "step": 18130
    },
    {
      "entropy": 5.709077787399292,
      "epoch": 1.6450471698113207,
      "grad_norm": 4.46875,
      "learning_rate": 0.00047322810389231333,
      "loss": 5.3962,
      "mean_token_accuracy": 0.1736616387963295,
      "num_tokens": 37639339.0,
      "step": 18135
    },
    {
      "entropy": 5.703421926498413,
      "epoch": 1.645500725689405,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004732127943253868,
      "loss": 5.4055,
      "mean_token_accuracy": 0.17203699797391891,
      "num_tokens": 37648836.0,
      "step": 18140
    },
    {
      "entropy": 5.708119487762451,
      "epoch": 1.645954281567489,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004731974806593693,
      "loss": 5.4406,
      "mean_token_accuracy": 0.16937143057584764,
      "num_tokens": 37658424.0,
      "step": 18145
    },
    {
      "entropy": 5.753741979598999,
      "epoch": 1.6464078374455733,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047318216289457727,
      "loss": 5.4577,
      "mean_token_accuracy": 0.16710167974233628,
      "num_tokens": 37669849.0,
      "step": 18150
    },
    {
      "entropy": 5.738954687118531,
      "epoch": 1.6468613933236576,
      "grad_norm": 4.1875,
      "learning_rate": 0.00047316684103132766,
      "loss": 5.4161,
      "mean_token_accuracy": 0.16822160929441451,
      "num_tokens": 37678987.0,
      "step": 18155
    },
    {
      "entropy": 5.796928930282593,
      "epoch": 1.6473149492017418,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004731515150699372,
      "loss": 5.4879,
      "mean_token_accuracy": 0.16710883006453514,
      "num_tokens": 37689692.0,
      "step": 18160
    },
    {
      "entropy": 5.743461751937867,
      "epoch": 1.6477685050798259,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047313618501072303,
      "loss": 5.6084,
      "mean_token_accuracy": 0.16274239718914033,
      "num_tokens": 37700092.0,
      "step": 18165
    },
    {
      "entropy": 5.739348602294922,
      "epoch": 1.64822206095791,
      "grad_norm": 7.21875,
      "learning_rate": 0.00047312085085400194,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16355093270540239,
      "num_tokens": 37710305.0,
      "step": 18170
    },
    {
      "entropy": 5.86072359085083,
      "epoch": 1.6486756168359942,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047310551260009126,
      "loss": 5.5986,
      "mean_token_accuracy": 0.1564181074500084,
      "num_tokens": 37720451.0,
      "step": 18175
    },
    {
      "entropy": 5.7927569389343265,
      "epoch": 1.6491291727140784,
      "grad_norm": 3.09375,
      "learning_rate": 0.000473090170249308,
      "loss": 5.4882,
      "mean_token_accuracy": 0.17014597952365876,
      "num_tokens": 37731952.0,
      "step": 18180
    },
    {
      "entropy": 5.701117467880249,
      "epoch": 1.6495827285921627,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004730748238019696,
      "loss": 5.4064,
      "mean_token_accuracy": 0.1605720967054367,
      "num_tokens": 37742201.0,
      "step": 18185
    },
    {
      "entropy": 5.731489753723144,
      "epoch": 1.6500362844702468,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047305947325839337,
      "loss": 5.458,
      "mean_token_accuracy": 0.1687504008412361,
      "num_tokens": 37753317.0,
      "step": 18190
    },
    {
      "entropy": 5.802850246429443,
      "epoch": 1.6504898403483308,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004730441186188968,
      "loss": 5.4856,
      "mean_token_accuracy": 0.16957197487354278,
      "num_tokens": 37763452.0,
      "step": 18195
    },
    {
      "entropy": 5.769397783279419,
      "epoch": 1.650943396226415,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047302875988379734,
      "loss": 5.3846,
      "mean_token_accuracy": 0.16978875994682313,
      "num_tokens": 37772804.0,
      "step": 18200
    },
    {
      "entropy": 5.758958196640014,
      "epoch": 1.6513969521044993,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047301339705341266,
      "loss": 5.4832,
      "mean_token_accuracy": 0.16031144857406615,
      "num_tokens": 37784441.0,
      "step": 18205
    },
    {
      "entropy": 5.76420226097107,
      "epoch": 1.6518505079825836,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047299803012806056,
      "loss": 5.4877,
      "mean_token_accuracy": 0.16719307601451874,
      "num_tokens": 37795103.0,
      "step": 18210
    },
    {
      "entropy": 5.723282766342163,
      "epoch": 1.6523040638606676,
      "grad_norm": 2.75,
      "learning_rate": 0.0004729826591080586,
      "loss": 5.481,
      "mean_token_accuracy": 0.17183628529310227,
      "num_tokens": 37806654.0,
      "step": 18215
    },
    {
      "entropy": 5.711847400665283,
      "epoch": 1.6527576197387517,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047296728399372483,
      "loss": 5.4948,
      "mean_token_accuracy": 0.16917998641729354,
      "num_tokens": 37816673.0,
      "step": 18220
    },
    {
      "entropy": 5.784402179718017,
      "epoch": 1.653211175616836,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047295190478537725,
      "loss": 5.4858,
      "mean_token_accuracy": 0.16785819679498673,
      "num_tokens": 37827389.0,
      "step": 18225
    },
    {
      "entropy": 5.858094692230225,
      "epoch": 1.6536647314949202,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047293652148333376,
      "loss": 5.5248,
      "mean_token_accuracy": 0.16161922365427017,
      "num_tokens": 37838087.0,
      "step": 18230
    },
    {
      "entropy": 5.642183828353882,
      "epoch": 1.6541182873730045,
      "grad_norm": 2.75,
      "learning_rate": 0.00047292113408791266,
      "loss": 5.394,
      "mean_token_accuracy": 0.17479628920555115,
      "num_tokens": 37848630.0,
      "step": 18235
    },
    {
      "entropy": 5.638126420974731,
      "epoch": 1.6545718432510885,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047290574259943195,
      "loss": 5.3846,
      "mean_token_accuracy": 0.18031271696090698,
      "num_tokens": 37858689.0,
      "step": 18240
    },
    {
      "entropy": 5.805227136611938,
      "epoch": 1.6550253991291726,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047289034701821013,
      "loss": 5.5293,
      "mean_token_accuracy": 0.1628846764564514,
      "num_tokens": 37869049.0,
      "step": 18245
    },
    {
      "entropy": 5.7344255447387695,
      "epoch": 1.6554789550072568,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047287494734456545,
      "loss": 5.4317,
      "mean_token_accuracy": 0.16685255467891694,
      "num_tokens": 37879495.0,
      "step": 18250
    },
    {
      "entropy": 5.748775148391724,
      "epoch": 1.655932510885341,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047285954357881646,
      "loss": 5.4957,
      "mean_token_accuracy": 0.16222003400325774,
      "num_tokens": 37889442.0,
      "step": 18255
    },
    {
      "entropy": 5.805187845230103,
      "epoch": 1.6563860667634254,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004728441357212817,
      "loss": 5.5853,
      "mean_token_accuracy": 0.15804096236824988,
      "num_tokens": 37900147.0,
      "step": 18260
    },
    {
      "entropy": 5.790038537979126,
      "epoch": 1.6568396226415094,
      "grad_norm": 5.625,
      "learning_rate": 0.0004728287237722798,
      "loss": 5.4444,
      "mean_token_accuracy": 0.16436990350484848,
      "num_tokens": 37910307.0,
      "step": 18265
    },
    {
      "entropy": 5.730356454849243,
      "epoch": 1.6572931785195935,
      "grad_norm": 4.125,
      "learning_rate": 0.0004728133077321295,
      "loss": 5.4141,
      "mean_token_accuracy": 0.1684657022356987,
      "num_tokens": 37920838.0,
      "step": 18270
    },
    {
      "entropy": 5.748407983779908,
      "epoch": 1.6577467343976777,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047279788760114954,
      "loss": 5.4867,
      "mean_token_accuracy": 0.17050281614065171,
      "num_tokens": 37930717.0,
      "step": 18275
    },
    {
      "entropy": 5.809717655181885,
      "epoch": 1.658200290275762,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004727824633796589,
      "loss": 5.5107,
      "mean_token_accuracy": 0.16238983124494552,
      "num_tokens": 37941254.0,
      "step": 18280
    },
    {
      "entropy": 5.758111047744751,
      "epoch": 1.6586538461538463,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004727670350679765,
      "loss": 5.4854,
      "mean_token_accuracy": 0.17322724908590317,
      "num_tokens": 37952132.0,
      "step": 18285
    },
    {
      "entropy": 5.695143365859986,
      "epoch": 1.6591074020319303,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004727516026664214,
      "loss": 5.4705,
      "mean_token_accuracy": 0.16906748414039613,
      "num_tokens": 37962026.0,
      "step": 18290
    },
    {
      "entropy": 5.716233825683593,
      "epoch": 1.6595609579100146,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004727361661753128,
      "loss": 5.3995,
      "mean_token_accuracy": 0.16777879893779754,
      "num_tokens": 37972911.0,
      "step": 18295
    },
    {
      "entropy": 5.764112520217895,
      "epoch": 1.6600145137880986,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004727207255949699,
      "loss": 5.4368,
      "mean_token_accuracy": 0.17217255234718323,
      "num_tokens": 37983577.0,
      "step": 18300
    },
    {
      "entropy": 5.784258604049683,
      "epoch": 1.6604680696661829,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047270528092571197,
      "loss": 5.5414,
      "mean_token_accuracy": 0.160717111825943,
      "num_tokens": 37994294.0,
      "step": 18305
    },
    {
      "entropy": 5.760765409469604,
      "epoch": 1.6609216255442671,
      "grad_norm": 2.125,
      "learning_rate": 0.0004726898321678584,
      "loss": 5.4913,
      "mean_token_accuracy": 0.17032980024814606,
      "num_tokens": 38005144.0,
      "step": 18310
    },
    {
      "entropy": 5.737387943267822,
      "epoch": 1.6613751814223512,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004726743793217288,
      "loss": 5.51,
      "mean_token_accuracy": 0.16345882415771484,
      "num_tokens": 38015277.0,
      "step": 18315
    },
    {
      "entropy": 5.782260608673096,
      "epoch": 1.6618287373004355,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047265892238764265,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16704344153404235,
      "num_tokens": 38027530.0,
      "step": 18320
    },
    {
      "entropy": 5.835880899429322,
      "epoch": 1.6622822931785195,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004726434613659196,
      "loss": 5.5403,
      "mean_token_accuracy": 0.16739441752433776,
      "num_tokens": 38036895.0,
      "step": 18325
    },
    {
      "entropy": 5.8575469017028805,
      "epoch": 1.6627358490566038,
      "grad_norm": 3.75,
      "learning_rate": 0.0004726279962568794,
      "loss": 5.5531,
      "mean_token_accuracy": 0.16381294429302215,
      "num_tokens": 38047001.0,
      "step": 18330
    },
    {
      "entropy": 5.757306241989136,
      "epoch": 1.663189404934688,
      "grad_norm": 4.25,
      "learning_rate": 0.0004726125270608419,
      "loss": 5.4349,
      "mean_token_accuracy": 0.16919412612915039,
      "num_tokens": 38056815.0,
      "step": 18335
    },
    {
      "entropy": 5.744669437408447,
      "epoch": 1.6636429608127723,
      "grad_norm": 5.0,
      "learning_rate": 0.00047259705377812703,
      "loss": 5.5091,
      "mean_token_accuracy": 0.1688440352678299,
      "num_tokens": 38067139.0,
      "step": 18340
    },
    {
      "entropy": 5.762575292587281,
      "epoch": 1.6640965166908563,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004725815764090546,
      "loss": 5.4398,
      "mean_token_accuracy": 0.17206551283597946,
      "num_tokens": 38076824.0,
      "step": 18345
    },
    {
      "entropy": 5.75336103439331,
      "epoch": 1.6645500725689404,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047256609495394494,
      "loss": 5.4575,
      "mean_token_accuracy": 0.16463080495595933,
      "num_tokens": 38087517.0,
      "step": 18350
    },
    {
      "entropy": 5.803236484527588,
      "epoch": 1.6650036284470247,
      "grad_norm": 3.640625,
      "learning_rate": 0.000472550609413118,
      "loss": 5.478,
      "mean_token_accuracy": 0.1626366913318634,
      "num_tokens": 38098499.0,
      "step": 18355
    },
    {
      "entropy": 5.7403350353240965,
      "epoch": 1.665457184325109,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004725351197868942,
      "loss": 5.462,
      "mean_token_accuracy": 0.17032012790441514,
      "num_tokens": 38108327.0,
      "step": 18360
    },
    {
      "entropy": 5.816367101669312,
      "epoch": 1.6659107402031932,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047251962607559373,
      "loss": 5.4687,
      "mean_token_accuracy": 0.16527373790740968,
      "num_tokens": 38118667.0,
      "step": 18365
    },
    {
      "entropy": 5.703820705413818,
      "epoch": 1.6663642960812772,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047250412827953717,
      "loss": 5.4364,
      "mean_token_accuracy": 0.16767489314079284,
      "num_tokens": 38128571.0,
      "step": 18370
    },
    {
      "entropy": 5.7619349479675295,
      "epoch": 1.6668178519593613,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004724886263990448,
      "loss": 5.5589,
      "mean_token_accuracy": 0.16263164281845094,
      "num_tokens": 38138921.0,
      "step": 18375
    },
    {
      "entropy": 5.865912914276123,
      "epoch": 1.6672714078374455,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004724731204344372,
      "loss": 5.5214,
      "mean_token_accuracy": 0.1615311473608017,
      "num_tokens": 38148858.0,
      "step": 18380
    },
    {
      "entropy": 5.69480414390564,
      "epoch": 1.6677249637155298,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047245761038603536,
      "loss": 5.4052,
      "mean_token_accuracy": 0.16928636729717256,
      "num_tokens": 38159107.0,
      "step": 18385
    },
    {
      "entropy": 5.776996850967407,
      "epoch": 1.668178519593614,
      "grad_norm": 5.0,
      "learning_rate": 0.00047244209625415975,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1626813068985939,
      "num_tokens": 38169228.0,
      "step": 18390
    },
    {
      "entropy": 5.822554254531861,
      "epoch": 1.6686320754716981,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004724265780391312,
      "loss": 5.4954,
      "mean_token_accuracy": 0.16880050748586656,
      "num_tokens": 38179175.0,
      "step": 18395
    },
    {
      "entropy": 5.745730686187744,
      "epoch": 1.6690856313497822,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004724110557412708,
      "loss": 5.4888,
      "mean_token_accuracy": 0.16905795484781266,
      "num_tokens": 38188537.0,
      "step": 18400
    },
    {
      "entropy": 5.708060932159424,
      "epoch": 1.6695391872278664,
      "grad_norm": 3.25,
      "learning_rate": 0.00047239552936089943,
      "loss": 5.4693,
      "mean_token_accuracy": 0.16173785030841828,
      "num_tokens": 38198972.0,
      "step": 18405
    },
    {
      "entropy": 5.71643853187561,
      "epoch": 1.6699927431059507,
      "grad_norm": 4.5,
      "learning_rate": 0.00047237999889833816,
      "loss": 5.3775,
      "mean_token_accuracy": 0.17242649346590042,
      "num_tokens": 38209472.0,
      "step": 18410
    },
    {
      "entropy": 5.627423715591431,
      "epoch": 1.670446298984035,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047236446435390826,
      "loss": 5.3854,
      "mean_token_accuracy": 0.17005249112844467,
      "num_tokens": 38219977.0,
      "step": 18415
    },
    {
      "entropy": 5.720183563232422,
      "epoch": 1.670899854862119,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047234892572793093,
      "loss": 5.4568,
      "mean_token_accuracy": 0.16811905652284623,
      "num_tokens": 38230492.0,
      "step": 18420
    },
    {
      "entropy": 5.711100339889526,
      "epoch": 1.671353410740203,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004723333830207275,
      "loss": 5.4479,
      "mean_token_accuracy": 0.16582875996828078,
      "num_tokens": 38240799.0,
      "step": 18425
    },
    {
      "entropy": 5.747720241546631,
      "epoch": 1.6718069666182873,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047231783623261954,
      "loss": 5.4184,
      "mean_token_accuracy": 0.1648837849497795,
      "num_tokens": 38251439.0,
      "step": 18430
    },
    {
      "entropy": 5.843761920928955,
      "epoch": 1.6722605224963716,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004723022853639283,
      "loss": 5.5686,
      "mean_token_accuracy": 0.16165921837091446,
      "num_tokens": 38261795.0,
      "step": 18435
    },
    {
      "entropy": 5.717576360702514,
      "epoch": 1.6727140783744558,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004722867304149756,
      "loss": 5.4093,
      "mean_token_accuracy": 0.16864618360996247,
      "num_tokens": 38272172.0,
      "step": 18440
    },
    {
      "entropy": 5.762063980102539,
      "epoch": 1.67316763425254,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004722711713860831,
      "loss": 5.5203,
      "mean_token_accuracy": 0.15789263546466828,
      "num_tokens": 38282829.0,
      "step": 18445
    },
    {
      "entropy": 5.83631157875061,
      "epoch": 1.673621190130624,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004722556082775724,
      "loss": 5.5485,
      "mean_token_accuracy": 0.1654006764292717,
      "num_tokens": 38293793.0,
      "step": 18450
    },
    {
      "entropy": 5.812488317489624,
      "epoch": 1.6740747460087082,
      "grad_norm": 3.125,
      "learning_rate": 0.0004722400410897655,
      "loss": 5.5082,
      "mean_token_accuracy": 0.16618854105472564,
      "num_tokens": 38302577.0,
      "step": 18455
    },
    {
      "entropy": 5.806685066223144,
      "epoch": 1.6745283018867925,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004722244698229842,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1667003244161606,
      "num_tokens": 38312571.0,
      "step": 18460
    },
    {
      "entropy": 5.759980821609497,
      "epoch": 1.6749818577648767,
      "grad_norm": 5.71875,
      "learning_rate": 0.00047220889447755073,
      "loss": 5.4076,
      "mean_token_accuracy": 0.17196781039237977,
      "num_tokens": 38323544.0,
      "step": 18465
    },
    {
      "entropy": 5.760677671432495,
      "epoch": 1.6754354136429608,
      "grad_norm": 4.875,
      "learning_rate": 0.0004721933150537869,
      "loss": 5.4878,
      "mean_token_accuracy": 0.16331418305635453,
      "num_tokens": 38334509.0,
      "step": 18470
    },
    {
      "entropy": 5.684898138046265,
      "epoch": 1.675888969521045,
      "grad_norm": 8.1875,
      "learning_rate": 0.00047217773155201526,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1678375095129013,
      "num_tokens": 38344747.0,
      "step": 18475
    },
    {
      "entropy": 5.777748394012451,
      "epoch": 1.676342525399129,
      "grad_norm": 3.0,
      "learning_rate": 0.00047216214397255776,
      "loss": 5.5154,
      "mean_token_accuracy": 0.16382718384265899,
      "num_tokens": 38355671.0,
      "step": 18480
    },
    {
      "entropy": 5.786406135559082,
      "epoch": 1.6767960812772134,
      "grad_norm": 3.734375,
      "learning_rate": 0.00047214655231573693,
      "loss": 5.4842,
      "mean_token_accuracy": 0.16804833859205245,
      "num_tokens": 38366368.0,
      "step": 18485
    },
    {
      "entropy": 5.80912299156189,
      "epoch": 1.6772496371552976,
      "grad_norm": 3.25,
      "learning_rate": 0.0004721309565818752,
      "loss": 5.575,
      "mean_token_accuracy": 0.16539790332317353,
      "num_tokens": 38377322.0,
      "step": 18490
    },
    {
      "entropy": 5.7668462753295895,
      "epoch": 1.6777031930333817,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047211535677129496,
      "loss": 5.5287,
      "mean_token_accuracy": 0.16482571065425872,
      "num_tokens": 38389204.0,
      "step": 18495
    },
    {
      "entropy": 5.740054416656494,
      "epoch": 1.678156748911466,
      "grad_norm": 4.5,
      "learning_rate": 0.00047209975288431893,
      "loss": 5.4739,
      "mean_token_accuracy": 0.16457287222146988,
      "num_tokens": 38399117.0,
      "step": 18500
    },
    {
      "entropy": 5.771087265014648,
      "epoch": 1.67861030478955,
      "grad_norm": 4.0625,
      "learning_rate": 0.00047208414492126986,
      "loss": 5.4589,
      "mean_token_accuracy": 0.16231805682182313,
      "num_tokens": 38410341.0,
      "step": 18505
    },
    {
      "entropy": 5.771722412109375,
      "epoch": 1.6790638606676342,
      "grad_norm": 5.375,
      "learning_rate": 0.00047206853288247043,
      "loss": 5.525,
      "mean_token_accuracy": 0.1600183479487896,
      "num_tokens": 38422742.0,
      "step": 18510
    },
    {
      "entropy": 5.766100549697876,
      "epoch": 1.6795174165457185,
      "grad_norm": 4.59375,
      "learning_rate": 0.00047205291676824345,
      "loss": 5.5121,
      "mean_token_accuracy": 0.16038536578416823,
      "num_tokens": 38433022.0,
      "step": 18515
    },
    {
      "entropy": 5.834670352935791,
      "epoch": 1.6799709724238028,
      "grad_norm": 3.109375,
      "learning_rate": 0.000472037296578912,
      "loss": 5.5126,
      "mean_token_accuracy": 0.1628789409995079,
      "num_tokens": 38444325.0,
      "step": 18520
    },
    {
      "entropy": 5.811626243591308,
      "epoch": 1.6804245283018868,
      "grad_norm": 2.953125,
      "learning_rate": 0.000472021672314799,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1709443137049675,
      "num_tokens": 38454576.0,
      "step": 18525
    },
    {
      "entropy": 5.769378614425659,
      "epoch": 1.6808780841799709,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047200604397622763,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1729914128780365,
      "num_tokens": 38465090.0,
      "step": 18530
    },
    {
      "entropy": 5.76339201927185,
      "epoch": 1.6813316400580551,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004719904115635211,
      "loss": 5.6518,
      "mean_token_accuracy": 0.15623801574110985,
      "num_tokens": 38477253.0,
      "step": 18535
    },
    {
      "entropy": 5.774687099456787,
      "epoch": 1.6817851959361394,
      "grad_norm": 3.375,
      "learning_rate": 0.0004719747750770026,
      "loss": 5.5944,
      "mean_token_accuracy": 0.16042157560586928,
      "num_tokens": 38488040.0,
      "step": 18540
    },
    {
      "entropy": 5.7426924228668215,
      "epoch": 1.6822387518142237,
      "grad_norm": 2.5,
      "learning_rate": 0.0004719591345169956,
      "loss": 5.4632,
      "mean_token_accuracy": 0.17070575058460236,
      "num_tokens": 38498016.0,
      "step": 18545
    },
    {
      "entropy": 5.8328639507293705,
      "epoch": 1.6826923076923077,
      "grad_norm": 3.625,
      "learning_rate": 0.00047194348988382345,
      "loss": 5.5555,
      "mean_token_accuracy": 0.1606586754322052,
      "num_tokens": 38508996.0,
      "step": 18550
    },
    {
      "entropy": 5.776587629318238,
      "epoch": 1.6831458635703918,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004719278411778097,
      "loss": 5.5115,
      "mean_token_accuracy": 0.1700441911816597,
      "num_tokens": 38519608.0,
      "step": 18555
    },
    {
      "entropy": 5.741469955444336,
      "epoch": 1.683599419448476,
      "grad_norm": 4.1875,
      "learning_rate": 0.00047191218839927817,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16484360098838807,
      "num_tokens": 38530737.0,
      "step": 18560
    },
    {
      "entropy": 5.759820652008057,
      "epoch": 1.6840529753265603,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004718965315485523,
      "loss": 5.4095,
      "mean_token_accuracy": 0.1667676404118538,
      "num_tokens": 38540925.0,
      "step": 18565
    },
    {
      "entropy": 5.755922937393189,
      "epoch": 1.6845065312046446,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004718808706259561,
      "loss": 5.4952,
      "mean_token_accuracy": 0.16058042347431184,
      "num_tokens": 38553176.0,
      "step": 18570
    },
    {
      "entropy": 5.794749927520752,
      "epoch": 1.6849600870827286,
      "grad_norm": 8.0,
      "learning_rate": 0.00047186520563181325,
      "loss": 5.5266,
      "mean_token_accuracy": 0.15906585603952408,
      "num_tokens": 38563394.0,
      "step": 18575
    },
    {
      "entropy": 5.828059720993042,
      "epoch": 1.6854136429608126,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004718495365664477,
      "loss": 5.5121,
      "mean_token_accuracy": 0.16675767302513123,
      "num_tokens": 38573243.0,
      "step": 18580
    },
    {
      "entropy": 5.794687032699585,
      "epoch": 1.685867198838897,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047183386343018364,
      "loss": 5.4951,
      "mean_token_accuracy": 0.1629287838935852,
      "num_tokens": 38584014.0,
      "step": 18585
    },
    {
      "entropy": 5.762561798095703,
      "epoch": 1.6863207547169812,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004718181862233451,
      "loss": 5.4426,
      "mean_token_accuracy": 0.17045405507087708,
      "num_tokens": 38594467.0,
      "step": 18590
    },
    {
      "entropy": 5.791234254837036,
      "epoch": 1.6867743105950654,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004718025049462563,
      "loss": 5.5054,
      "mean_token_accuracy": 0.16608012914657594,
      "num_tokens": 38604407.0,
      "step": 18595
    },
    {
      "entropy": 5.819661426544189,
      "epoch": 1.6872278664731495,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004717868195992416,
      "loss": 5.5217,
      "mean_token_accuracy": 0.17121877372264863,
      "num_tokens": 38614508.0,
      "step": 18600
    },
    {
      "entropy": 5.888324213027954,
      "epoch": 1.6876814223512335,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004717711301826253,
      "loss": 5.5307,
      "mean_token_accuracy": 0.16247211992740632,
      "num_tokens": 38626244.0,
      "step": 18605
    },
    {
      "entropy": 5.728854322433472,
      "epoch": 1.6881349782293178,
      "grad_norm": 8.625,
      "learning_rate": 0.00047175543669673186,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16663504242897034,
      "num_tokens": 38636694.0,
      "step": 18610
    },
    {
      "entropy": 5.826562213897705,
      "epoch": 1.688588534107402,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047173973914188587,
      "loss": 5.5712,
      "mean_token_accuracy": 0.16274236291646957,
      "num_tokens": 38647414.0,
      "step": 18615
    },
    {
      "entropy": 5.7988972663879395,
      "epoch": 1.6890420899854863,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047172403751841196,
      "loss": 5.4862,
      "mean_token_accuracy": 0.16609843969345092,
      "num_tokens": 38657425.0,
      "step": 18620
    },
    {
      "entropy": 5.789045238494873,
      "epoch": 1.6894956458635704,
      "grad_norm": 5.0,
      "learning_rate": 0.00047170833182663485,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1667177453637123,
      "num_tokens": 38667751.0,
      "step": 18625
    },
    {
      "entropy": 5.686364078521729,
      "epoch": 1.6899492017416544,
      "grad_norm": 4.875,
      "learning_rate": 0.0004716926220668793,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1697746306657791,
      "num_tokens": 38676918.0,
      "step": 18630
    },
    {
      "entropy": 5.742925643920898,
      "epoch": 1.6904027576197387,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047167690823947015,
      "loss": 5.4724,
      "mean_token_accuracy": 0.17273928374052047,
      "num_tokens": 38685887.0,
      "step": 18635
    },
    {
      "entropy": 5.741593551635742,
      "epoch": 1.690856313497823,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004716611903447325,
      "loss": 5.4663,
      "mean_token_accuracy": 0.17263988256454468,
      "num_tokens": 38696179.0,
      "step": 18640
    },
    {
      "entropy": 5.756220960617066,
      "epoch": 1.6913098693759072,
      "grad_norm": 4.375,
      "learning_rate": 0.0004716454683829912,
      "loss": 5.4532,
      "mean_token_accuracy": 0.16362424790859223,
      "num_tokens": 38705777.0,
      "step": 18645
    },
    {
      "entropy": 5.739682006835937,
      "epoch": 1.6917634252539913,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047162974235457163,
      "loss": 5.4952,
      "mean_token_accuracy": 0.16620824635028839,
      "num_tokens": 38717709.0,
      "step": 18650
    },
    {
      "entropy": 5.787838888168335,
      "epoch": 1.6922169811320755,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004716140122597988,
      "loss": 5.4627,
      "mean_token_accuracy": 0.1710629478096962,
      "num_tokens": 38729517.0,
      "step": 18655
    },
    {
      "entropy": 5.842197322845459,
      "epoch": 1.6926705370101596,
      "grad_norm": 2.75,
      "learning_rate": 0.0004715982780989981,
      "loss": 5.5021,
      "mean_token_accuracy": 0.16779959052801133,
      "num_tokens": 38740878.0,
      "step": 18660
    },
    {
      "entropy": 5.775685358047485,
      "epoch": 1.6931240928882438,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047158253987249487,
      "loss": 5.4193,
      "mean_token_accuracy": 0.1666555657982826,
      "num_tokens": 38751291.0,
      "step": 18665
    },
    {
      "entropy": 5.663282155990601,
      "epoch": 1.693577648766328,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047156679758061465,
      "loss": 5.3928,
      "mean_token_accuracy": 0.17812413573265076,
      "num_tokens": 38760940.0,
      "step": 18670
    },
    {
      "entropy": 5.720236539840698,
      "epoch": 1.6940312046444121,
      "grad_norm": 5.125,
      "learning_rate": 0.00047155105122368297,
      "loss": 5.4017,
      "mean_token_accuracy": 0.17299514412879943,
      "num_tokens": 38770240.0,
      "step": 18675
    },
    {
      "entropy": 5.743992614746094,
      "epoch": 1.6944847605224964,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004715353008020255,
      "loss": 5.4295,
      "mean_token_accuracy": 0.17061990797519683,
      "num_tokens": 38780452.0,
      "step": 18680
    },
    {
      "entropy": 5.785730934143066,
      "epoch": 1.6949383164005805,
      "grad_norm": 3.046875,
      "learning_rate": 0.00047151954631596794,
      "loss": 5.5005,
      "mean_token_accuracy": 0.16809303164482117,
      "num_tokens": 38790763.0,
      "step": 18685
    },
    {
      "entropy": 5.772569751739502,
      "epoch": 1.6953918722786647,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047150378776583603,
      "loss": 5.3932,
      "mean_token_accuracy": 0.17547277808189393,
      "num_tokens": 38802164.0,
      "step": 18690
    },
    {
      "entropy": 5.751386404037476,
      "epoch": 1.695845428156749,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004714880251519557,
      "loss": 5.4693,
      "mean_token_accuracy": 0.16368554532527924,
      "num_tokens": 38812901.0,
      "step": 18695
    },
    {
      "entropy": 5.7535947322845455,
      "epoch": 1.6962989840348333,
      "grad_norm": 3.515625,
      "learning_rate": 0.000471472258474653,
      "loss": 5.4606,
      "mean_token_accuracy": 0.1637917459011078,
      "num_tokens": 38823305.0,
      "step": 18700
    },
    {
      "entropy": 5.752395868301392,
      "epoch": 1.6967525399129173,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047145648773425384,
      "loss": 5.4251,
      "mean_token_accuracy": 0.16004481315612792,
      "num_tokens": 38833602.0,
      "step": 18705
    },
    {
      "entropy": 5.77405047416687,
      "epoch": 1.6972060957910013,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047144071293108456,
      "loss": 5.5233,
      "mean_token_accuracy": 0.16370468139648436,
      "num_tokens": 38843950.0,
      "step": 18710
    },
    {
      "entropy": 5.744354391098023,
      "epoch": 1.6976596516690856,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004714249340654712,
      "loss": 5.4212,
      "mean_token_accuracy": 0.1678694188594818,
      "num_tokens": 38853753.0,
      "step": 18715
    },
    {
      "entropy": 5.682321310043335,
      "epoch": 1.6981132075471699,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004714091511377402,
      "loss": 5.3767,
      "mean_token_accuracy": 0.16418073028326036,
      "num_tokens": 38863394.0,
      "step": 18720
    },
    {
      "entropy": 5.752600240707397,
      "epoch": 1.6985667634252541,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047139336414821784,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16667638272047042,
      "num_tokens": 38872972.0,
      "step": 18725
    },
    {
      "entropy": 5.760299730300903,
      "epoch": 1.6990203193033382,
      "grad_norm": 5.125,
      "learning_rate": 0.00047137757309723073,
      "loss": 5.4611,
      "mean_token_accuracy": 0.1737182155251503,
      "num_tokens": 38883251.0,
      "step": 18730
    },
    {
      "entropy": 5.6902174949646,
      "epoch": 1.6994738751814222,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004713617779851053,
      "loss": 5.464,
      "mean_token_accuracy": 0.1648678883910179,
      "num_tokens": 38893274.0,
      "step": 18735
    },
    {
      "entropy": 5.803202056884766,
      "epoch": 1.6999274310595065,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004713459788121683,
      "loss": 5.5571,
      "mean_token_accuracy": 0.16598086953163146,
      "num_tokens": 38903870.0,
      "step": 18740
    },
    {
      "entropy": 5.826020574569702,
      "epoch": 1.7003809869375908,
      "grad_norm": 2.25,
      "learning_rate": 0.0004713301755787464,
      "loss": 5.5189,
      "mean_token_accuracy": 0.1608530730009079,
      "num_tokens": 38914921.0,
      "step": 18745
    },
    {
      "entropy": 5.807393169403076,
      "epoch": 1.700834542815675,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004713143682851665,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16852097511291503,
      "num_tokens": 38924630.0,
      "step": 18750
    },
    {
      "entropy": 5.725153207778931,
      "epoch": 1.701288098693759,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004712985569317553,
      "loss": 5.4458,
      "mean_token_accuracy": 0.169877228140831,
      "num_tokens": 38934926.0,
      "step": 18755
    },
    {
      "entropy": 5.807769918441773,
      "epoch": 1.7017416545718431,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004712827415188401,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1566988930106163,
      "num_tokens": 38944513.0,
      "step": 18760
    },
    {
      "entropy": 5.800018930435181,
      "epoch": 1.7021952104499274,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004712669220467476,
      "loss": 5.4674,
      "mean_token_accuracy": 0.16440123915672303,
      "num_tokens": 38955918.0,
      "step": 18765
    },
    {
      "entropy": 5.810863447189331,
      "epoch": 1.7026487663280117,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047125109851580533,
      "loss": 5.4681,
      "mean_token_accuracy": 0.16796416491270066,
      "num_tokens": 38966221.0,
      "step": 18770
    },
    {
      "entropy": 5.7775510311126705,
      "epoch": 1.703102322206096,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047123527092634025,
      "loss": 5.4345,
      "mean_token_accuracy": 0.16634514033794404,
      "num_tokens": 38976463.0,
      "step": 18775
    },
    {
      "entropy": 5.689714622497559,
      "epoch": 1.70355587808418,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004712194392786797,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17269947081804277,
      "num_tokens": 38986375.0,
      "step": 18780
    },
    {
      "entropy": 5.843862009048462,
      "epoch": 1.704009433962264,
      "grad_norm": 5.40625,
      "learning_rate": 0.00047120360357315127,
      "loss": 5.5307,
      "mean_token_accuracy": 0.16342256516218184,
      "num_tokens": 38996116.0,
      "step": 18785
    },
    {
      "entropy": 5.758773517608643,
      "epoch": 1.7044629898403483,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047118776381008216,
      "loss": 5.4622,
      "mean_token_accuracy": 0.1710265025496483,
      "num_tokens": 39005846.0,
      "step": 18790
    },
    {
      "entropy": 5.767974805831909,
      "epoch": 1.7049165457184325,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047117191998980026,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1628724679350853,
      "num_tokens": 39016744.0,
      "step": 18795
    },
    {
      "entropy": 5.7719635486602785,
      "epoch": 1.7053701015965168,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047115607211263295,
      "loss": 5.5299,
      "mean_token_accuracy": 0.16403818428516387,
      "num_tokens": 39026725.0,
      "step": 18800
    },
    {
      "entropy": 5.816298389434815,
      "epoch": 1.7058236574746009,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047114022017890814,
      "loss": 5.4784,
      "mean_token_accuracy": 0.16420698761940003,
      "num_tokens": 39036851.0,
      "step": 18805
    },
    {
      "entropy": 5.758962392807007,
      "epoch": 1.706277213352685,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004711243641889536,
      "loss": 5.4015,
      "mean_token_accuracy": 0.16516944915056228,
      "num_tokens": 39047179.0,
      "step": 18810
    },
    {
      "entropy": 5.733838987350464,
      "epoch": 1.7067307692307692,
      "grad_norm": 3.296875,
      "learning_rate": 0.00047110850414309714,
      "loss": 5.3147,
      "mean_token_accuracy": 0.17749351859092713,
      "num_tokens": 39057758.0,
      "step": 18815
    },
    {
      "entropy": 5.72713189125061,
      "epoch": 1.7071843251088534,
      "grad_norm": 3.6875,
      "learning_rate": 0.00047109264004166696,
      "loss": 5.5317,
      "mean_token_accuracy": 0.16293642669916153,
      "num_tokens": 39068142.0,
      "step": 18820
    },
    {
      "entropy": 5.7479345321655275,
      "epoch": 1.7076378809869377,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004710767718849909,
      "loss": 5.4392,
      "mean_token_accuracy": 0.1639469027519226,
      "num_tokens": 39078157.0,
      "step": 18825
    },
    {
      "entropy": 5.764247083663941,
      "epoch": 1.7080914368650217,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004710608996733973,
      "loss": 5.4722,
      "mean_token_accuracy": 0.17323556691408157,
      "num_tokens": 39089090.0,
      "step": 18830
    },
    {
      "entropy": 5.74896936416626,
      "epoch": 1.708544992743106,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047104502340721433,
      "loss": 5.5094,
      "mean_token_accuracy": 0.15809889733791352,
      "num_tokens": 39099079.0,
      "step": 18835
    },
    {
      "entropy": 5.77755036354065,
      "epoch": 1.70899854862119,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047102914308677033,
      "loss": 5.4677,
      "mean_token_accuracy": 0.16553575098514556,
      "num_tokens": 39109927.0,
      "step": 18840
    },
    {
      "entropy": 5.765803050994873,
      "epoch": 1.7094521044992743,
      "grad_norm": 2.5,
      "learning_rate": 0.00047101325871239363,
      "loss": 5.4617,
      "mean_token_accuracy": 0.16879944503307343,
      "num_tokens": 39119806.0,
      "step": 18845
    },
    {
      "entropy": 5.808726835250854,
      "epoch": 1.7099056603773586,
      "grad_norm": 2.25,
      "learning_rate": 0.0004709973702844128,
      "loss": 5.5465,
      "mean_token_accuracy": 0.16456539928913116,
      "num_tokens": 39130147.0,
      "step": 18850
    },
    {
      "entropy": 5.728408718109131,
      "epoch": 1.7103592162554426,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047098147780315645,
      "loss": 5.3883,
      "mean_token_accuracy": 0.1734140157699585,
      "num_tokens": 39139937.0,
      "step": 18855
    },
    {
      "entropy": 5.789205884933471,
      "epoch": 1.710812772133527,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004709655812689532,
      "loss": 5.4902,
      "mean_token_accuracy": 0.1644870549440384,
      "num_tokens": 39149977.0,
      "step": 18860
    },
    {
      "entropy": 5.791285037994385,
      "epoch": 1.711266328011611,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004709496806821318,
      "loss": 5.4982,
      "mean_token_accuracy": 0.16547591388225555,
      "num_tokens": 39161076.0,
      "step": 18865
    },
    {
      "entropy": 5.791528177261353,
      "epoch": 1.7117198838896952,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047093377604302106,
      "loss": 5.5169,
      "mean_token_accuracy": 0.16235332936048508,
      "num_tokens": 39170776.0,
      "step": 18870
    },
    {
      "entropy": 5.761791515350342,
      "epoch": 1.7121734397677795,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004709178673519499,
      "loss": 5.4822,
      "mean_token_accuracy": 0.16460847109556198,
      "num_tokens": 39181134.0,
      "step": 18875
    },
    {
      "entropy": 5.767756843566895,
      "epoch": 1.7126269956458637,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004709019546092473,
      "loss": 5.5878,
      "mean_token_accuracy": 0.15967290997505187,
      "num_tokens": 39193331.0,
      "step": 18880
    },
    {
      "entropy": 5.873799085617065,
      "epoch": 1.7130805515239478,
      "grad_norm": 8.4375,
      "learning_rate": 0.00047088603781524237,
      "loss": 5.5445,
      "mean_token_accuracy": 0.1599661335349083,
      "num_tokens": 39204372.0,
      "step": 18885
    },
    {
      "entropy": 5.8106041431427,
      "epoch": 1.7135341074020318,
      "grad_norm": 3.125,
      "learning_rate": 0.0004708701169702642,
      "loss": 5.4603,
      "mean_token_accuracy": 0.17246332615613938,
      "num_tokens": 39214725.0,
      "step": 18890
    },
    {
      "entropy": 5.6511091709136965,
      "epoch": 1.713987663280116,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004708541920746422,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16875732690095901,
      "num_tokens": 39225119.0,
      "step": 18895
    },
    {
      "entropy": 5.746160840988159,
      "epoch": 1.7144412191582004,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004708382631287055,
      "loss": 5.4105,
      "mean_token_accuracy": 0.168777796626091,
      "num_tokens": 39235021.0,
      "step": 18900
    },
    {
      "entropy": 5.811975812911987,
      "epoch": 1.7148947750362846,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004708223301327837,
      "loss": 5.5815,
      "mean_token_accuracy": 0.1602236047387123,
      "num_tokens": 39245830.0,
      "step": 18905
    },
    {
      "entropy": 5.815212869644165,
      "epoch": 1.7153483309143687,
      "grad_norm": 4.71875,
      "learning_rate": 0.00047080639308720617,
      "loss": 5.5638,
      "mean_token_accuracy": 0.16522229164838792,
      "num_tokens": 39255264.0,
      "step": 18910
    },
    {
      "entropy": 5.766363430023193,
      "epoch": 1.7158018867924527,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047079045199230254,
      "loss": 5.4461,
      "mean_token_accuracy": 0.16678517907857895,
      "num_tokens": 39264966.0,
      "step": 18915
    },
    {
      "entropy": 5.771310138702392,
      "epoch": 1.716255442670537,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047077450684840247,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1624810129404068,
      "num_tokens": 39275082.0,
      "step": 18920
    },
    {
      "entropy": 5.812801694869995,
      "epoch": 1.7167089985486212,
      "grad_norm": 2.875,
      "learning_rate": 0.0004707585576558357,
      "loss": 5.5921,
      "mean_token_accuracy": 0.1586887314915657,
      "num_tokens": 39285426.0,
      "step": 18925
    },
    {
      "entropy": 5.906033658981324,
      "epoch": 1.7171625544267055,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004707426044149321,
      "loss": 5.5824,
      "mean_token_accuracy": 0.1601039484143257,
      "num_tokens": 39296633.0,
      "step": 18930
    },
    {
      "entropy": 5.734673976898193,
      "epoch": 1.7176161103047896,
      "grad_norm": 3.609375,
      "learning_rate": 0.00047072664712602153,
      "loss": 5.375,
      "mean_token_accuracy": 0.1781951978802681,
      "num_tokens": 39306775.0,
      "step": 18935
    },
    {
      "entropy": 5.759129047393799,
      "epoch": 1.7180696661828736,
      "grad_norm": 3.515625,
      "learning_rate": 0.000470710685789434,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1604754850268364,
      "num_tokens": 39317917.0,
      "step": 18940
    },
    {
      "entropy": 5.78073296546936,
      "epoch": 1.7185232220609579,
      "grad_norm": 3.671875,
      "learning_rate": 0.00047069472040549965,
      "loss": 5.4976,
      "mean_token_accuracy": 0.1627676621079445,
      "num_tokens": 39328387.0,
      "step": 18945
    },
    {
      "entropy": 5.670525121688843,
      "epoch": 1.7189767779390421,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004706787509745486,
      "loss": 5.3495,
      "mean_token_accuracy": 0.17618516236543655,
      "num_tokens": 39337639.0,
      "step": 18950
    },
    {
      "entropy": 5.740663862228393,
      "epoch": 1.7194303338171264,
      "grad_norm": 2.625,
      "learning_rate": 0.00047066277749691115,
      "loss": 5.4805,
      "mean_token_accuracy": 0.16927317827939986,
      "num_tokens": 39347971.0,
      "step": 18955
    },
    {
      "entropy": 5.785571479797364,
      "epoch": 1.7198838896952104,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004706467999729175,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16604194045066833,
      "num_tokens": 39358038.0,
      "step": 18960
    },
    {
      "entropy": 5.792507028579712,
      "epoch": 1.7203374455732945,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004706308184028983,
      "loss": 5.4968,
      "mean_token_accuracy": 0.1664569318294525,
      "num_tokens": 39367834.0,
      "step": 18965
    },
    {
      "entropy": 5.748879098892212,
      "epoch": 1.7207910014513788,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004706148327871839,
      "loss": 5.5044,
      "mean_token_accuracy": 0.16748203486204147,
      "num_tokens": 39378924.0,
      "step": 18970
    },
    {
      "entropy": 5.836343717575073,
      "epoch": 1.721244557329463,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047059884312610487,
      "loss": 5.6108,
      "mean_token_accuracy": 0.15930540934205056,
      "num_tokens": 39390082.0,
      "step": 18975
    },
    {
      "entropy": 5.792299461364746,
      "epoch": 1.7216981132075473,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047058284941999185,
      "loss": 5.4966,
      "mean_token_accuracy": 0.16593593955039979,
      "num_tokens": 39400485.0,
      "step": 18980
    },
    {
      "entropy": 5.824423694610596,
      "epoch": 1.7221516690856313,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004705668516691758,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16830268800258635,
      "num_tokens": 39411601.0,
      "step": 18985
    },
    {
      "entropy": 5.779353475570678,
      "epoch": 1.7226052249637154,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047055084987398735,
      "loss": 5.4117,
      "mean_token_accuracy": 0.17091877162456512,
      "num_tokens": 39421625.0,
      "step": 18990
    },
    {
      "entropy": 5.710326910018921,
      "epoch": 1.7230587808417996,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004705348440347575,
      "loss": 5.4422,
      "mean_token_accuracy": 0.16403636634349822,
      "num_tokens": 39432033.0,
      "step": 18995
    },
    {
      "entropy": 5.726392364501953,
      "epoch": 1.723512336719884,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047051883415181724,
      "loss": 5.4635,
      "mean_token_accuracy": 0.1653407871723175,
      "num_tokens": 39442514.0,
      "step": 19000
    },
    {
      "entropy": 5.759695625305175,
      "epoch": 1.7239658925979682,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004705028202254977,
      "loss": 5.362,
      "mean_token_accuracy": 0.17458819150924682,
      "num_tokens": 39452497.0,
      "step": 19005
    },
    {
      "entropy": 5.729509162902832,
      "epoch": 1.7244194484760522,
      "grad_norm": 8.4375,
      "learning_rate": 0.00047048680225612996,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16845736056566238,
      "num_tokens": 39462173.0,
      "step": 19010
    },
    {
      "entropy": 5.754267835617066,
      "epoch": 1.7248730043541363,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004704707802440454,
      "loss": 5.5347,
      "mean_token_accuracy": 0.17139464616775513,
      "num_tokens": 39473123.0,
      "step": 19015
    },
    {
      "entropy": 5.773678016662598,
      "epoch": 1.7253265602322205,
      "grad_norm": 3.78125,
      "learning_rate": 0.00047045475418957524,
      "loss": 5.4836,
      "mean_token_accuracy": 0.15831948667764664,
      "num_tokens": 39484142.0,
      "step": 19020
    },
    {
      "entropy": 5.789837121963501,
      "epoch": 1.7257801161103048,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004704387240930509,
      "loss": 5.5037,
      "mean_token_accuracy": 0.16140259504318238,
      "num_tokens": 39494926.0,
      "step": 19025
    },
    {
      "entropy": 5.768725156784058,
      "epoch": 1.726233671988389,
      "grad_norm": 4.875,
      "learning_rate": 0.0004704226899548041,
      "loss": 5.4506,
      "mean_token_accuracy": 0.16877688765525817,
      "num_tokens": 39505153.0,
      "step": 19030
    },
    {
      "entropy": 5.698583459854126,
      "epoch": 1.726687227866473,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047040665177516616,
      "loss": 5.436,
      "mean_token_accuracy": 0.17083770185709,
      "num_tokens": 39515905.0,
      "step": 19035
    },
    {
      "entropy": 5.727232456207275,
      "epoch": 1.7271407837445574,
      "grad_norm": 3.921875,
      "learning_rate": 0.00047039060955446886,
      "loss": 5.4481,
      "mean_token_accuracy": 0.16622323244810105,
      "num_tokens": 39526762.0,
      "step": 19040
    },
    {
      "entropy": 5.722050905227661,
      "epoch": 1.7275943396226414,
      "grad_norm": 5.1875,
      "learning_rate": 0.000470374563293044,
      "loss": 5.4351,
      "mean_token_accuracy": 0.16422508805990219,
      "num_tokens": 39537340.0,
      "step": 19045
    },
    {
      "entropy": 5.733491039276123,
      "epoch": 1.7280478955007257,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004703585129912234,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16749505400657655,
      "num_tokens": 39547144.0,
      "step": 19050
    },
    {
      "entropy": 5.845774698257446,
      "epoch": 1.72850145137881,
      "grad_norm": 4.8125,
      "learning_rate": 0.00047034245864933894,
      "loss": 5.5708,
      "mean_token_accuracy": 0.16054221540689467,
      "num_tokens": 39557707.0,
      "step": 19055
    },
    {
      "entropy": 5.8201515674591064,
      "epoch": 1.728955007256894,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047032640026772274,
      "loss": 5.5358,
      "mean_token_accuracy": 0.1609910845756531,
      "num_tokens": 39567629.0,
      "step": 19060
    },
    {
      "entropy": 5.743157863616943,
      "epoch": 1.7294085631349783,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004703103378467067,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16755575239658355,
      "num_tokens": 39577918.0,
      "step": 19065
    },
    {
      "entropy": 5.741540145874024,
      "epoch": 1.7298621190130623,
      "grad_norm": 2.125,
      "learning_rate": 0.00047029427138662314,
      "loss": 5.5189,
      "mean_token_accuracy": 0.1657979443669319,
      "num_tokens": 39587838.0,
      "step": 19070
    },
    {
      "entropy": 5.696336174011231,
      "epoch": 1.7303156748911466,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004702782008878042,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17529295086860658,
      "num_tokens": 39597353.0,
      "step": 19075
    },
    {
      "entropy": 5.739788770675659,
      "epoch": 1.7307692307692308,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004702621263505824,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1685292512178421,
      "num_tokens": 39607349.0,
      "step": 19080
    },
    {
      "entropy": 5.793078947067261,
      "epoch": 1.731222786647315,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004702460477752901,
      "loss": 5.5017,
      "mean_token_accuracy": 0.16354060918092728,
      "num_tokens": 39617810.0,
      "step": 19085
    },
    {
      "entropy": 5.744883012771607,
      "epoch": 1.7316763425253991,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004702299651622597,
      "loss": 5.3822,
      "mean_token_accuracy": 0.1752336025238037,
      "num_tokens": 39626805.0,
      "step": 19090
    },
    {
      "entropy": 5.691711902618408,
      "epoch": 1.7321298984034832,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004702138785118239,
      "loss": 5.4313,
      "mean_token_accuracy": 0.1728903591632843,
      "num_tokens": 39637839.0,
      "step": 19095
    },
    {
      "entropy": 5.6900399684906,
      "epoch": 1.7325834542815675,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004701977878243153,
      "loss": 5.4254,
      "mean_token_accuracy": 0.16588961482048034,
      "num_tokens": 39648697.0,
      "step": 19100
    },
    {
      "entropy": 5.662890291213989,
      "epoch": 1.7330370101596517,
      "grad_norm": 2.625,
      "learning_rate": 0.0004701816931000668,
      "loss": 5.4195,
      "mean_token_accuracy": 0.171548031270504,
      "num_tokens": 39658154.0,
      "step": 19105
    },
    {
      "entropy": 5.781878852844239,
      "epoch": 1.733490566037736,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047016559433941106,
      "loss": 5.4933,
      "mean_token_accuracy": 0.16606540679931642,
      "num_tokens": 39667939.0,
      "step": 19110
    },
    {
      "entropy": 5.701201009750366,
      "epoch": 1.73394412191582,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004701494915426812,
      "loss": 5.3891,
      "mean_token_accuracy": 0.16641004383563995,
      "num_tokens": 39678289.0,
      "step": 19115
    },
    {
      "entropy": 5.749450778961181,
      "epoch": 1.734397677793904,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047013338471021,
      "loss": 5.4543,
      "mean_token_accuracy": 0.1658313527703285,
      "num_tokens": 39689901.0,
      "step": 19120
    },
    {
      "entropy": 5.821782970428467,
      "epoch": 1.7348512336719883,
      "grad_norm": 2.0,
      "learning_rate": 0.00047011727384233074,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1703617826104164,
      "num_tokens": 39700831.0,
      "step": 19125
    },
    {
      "entropy": 5.765407276153565,
      "epoch": 1.7353047895500726,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004701011589393765,
      "loss": 5.5122,
      "mean_token_accuracy": 0.16169517189264299,
      "num_tokens": 39711794.0,
      "step": 19130
    },
    {
      "entropy": 5.6896649360656735,
      "epoch": 1.7357583454281569,
      "grad_norm": 2.9375,
      "learning_rate": 0.00047008504000168054,
      "loss": 5.4403,
      "mean_token_accuracy": 0.16966246515512468,
      "num_tokens": 39721889.0,
      "step": 19135
    },
    {
      "entropy": 5.711557245254516,
      "epoch": 1.736211901306241,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004700689170295763,
      "loss": 5.5145,
      "mean_token_accuracy": 0.1658303052186966,
      "num_tokens": 39732562.0,
      "step": 19140
    },
    {
      "entropy": 5.848415899276733,
      "epoch": 1.736665457184325,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047005279002339717,
      "loss": 5.4573,
      "mean_token_accuracy": 0.16699902415275575,
      "num_tokens": 39742286.0,
      "step": 19145
    },
    {
      "entropy": 5.7573729991912845,
      "epoch": 1.7371190130624092,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047003665898347663,
      "loss": 5.4735,
      "mean_token_accuracy": 0.16764807552099228,
      "num_tokens": 39754336.0,
      "step": 19150
    },
    {
      "entropy": 5.6658483982086185,
      "epoch": 1.7375725689404935,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004700205239101483,
      "loss": 5.3773,
      "mean_token_accuracy": 0.17401509881019592,
      "num_tokens": 39763545.0,
      "step": 19155
    },
    {
      "entropy": 5.802091598510742,
      "epoch": 1.7380261248185778,
      "grad_norm": 4.375,
      "learning_rate": 0.0004700043848037458,
      "loss": 5.5382,
      "mean_token_accuracy": 0.15884328484535218,
      "num_tokens": 39774536.0,
      "step": 19160
    },
    {
      "entropy": 5.85611743927002,
      "epoch": 1.7384796806966618,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004699882416646029,
      "loss": 5.4803,
      "mean_token_accuracy": 0.16605832874774934,
      "num_tokens": 39785377.0,
      "step": 19165
    },
    {
      "entropy": 5.779321098327637,
      "epoch": 1.7389332365747459,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004699720944930535,
      "loss": 5.5493,
      "mean_token_accuracy": 0.15396990329027177,
      "num_tokens": 39796692.0,
      "step": 19170
    },
    {
      "entropy": 5.681128311157226,
      "epoch": 1.7393867924528301,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046995594328943147,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16378923654556274,
      "num_tokens": 39807927.0,
      "step": 19175
    },
    {
      "entropy": 5.86321988105774,
      "epoch": 1.7398403483309144,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004699397880540709,
      "loss": 5.6302,
      "mean_token_accuracy": 0.1632505550980568,
      "num_tokens": 39817617.0,
      "step": 19180
    },
    {
      "entropy": 5.7905419826507565,
      "epoch": 1.7402939042089987,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004699236287873058,
      "loss": 5.3775,
      "mean_token_accuracy": 0.17393258064985276,
      "num_tokens": 39827092.0,
      "step": 19185
    },
    {
      "entropy": 5.7287362575531,
      "epoch": 1.7407474600870827,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004699074654894705,
      "loss": 5.4269,
      "mean_token_accuracy": 0.168648599088192,
      "num_tokens": 39837378.0,
      "step": 19190
    },
    {
      "entropy": 5.766827583312988,
      "epoch": 1.7412010159651667,
      "grad_norm": 2.734375,
      "learning_rate": 0.000469891298160899,
      "loss": 5.4293,
      "mean_token_accuracy": 0.16253943592309952,
      "num_tokens": 39848191.0,
      "step": 19195
    },
    {
      "entropy": 5.714908218383789,
      "epoch": 1.741654571843251,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046987512680192586,
      "loss": 5.4214,
      "mean_token_accuracy": 0.16707662791013717,
      "num_tokens": 39857886.0,
      "step": 19200
    },
    {
      "entropy": 5.799066734313965,
      "epoch": 1.7421081277213353,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046985895141288546,
      "loss": 5.5331,
      "mean_token_accuracy": 0.16336908787488938,
      "num_tokens": 39868348.0,
      "step": 19205
    },
    {
      "entropy": 5.825392961502075,
      "epoch": 1.7425616835994195,
      "grad_norm": 4.0625,
      "learning_rate": 0.00046984277199411224,
      "loss": 5.5336,
      "mean_token_accuracy": 0.16452379375696183,
      "num_tokens": 39879129.0,
      "step": 19210
    },
    {
      "entropy": 5.726678276062012,
      "epoch": 1.7430152394775036,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046982658854594083,
      "loss": 5.4232,
      "mean_token_accuracy": 0.17032379806041717,
      "num_tokens": 39890239.0,
      "step": 19215
    },
    {
      "entropy": 5.670359420776367,
      "epoch": 1.7434687953555879,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046981040106870595,
      "loss": 5.3313,
      "mean_token_accuracy": 0.17865807861089705,
      "num_tokens": 39900907.0,
      "step": 19220
    },
    {
      "entropy": 5.757753133773804,
      "epoch": 1.743922351233672,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004697942095627423,
      "loss": 5.4686,
      "mean_token_accuracy": 0.16777524948120118,
      "num_tokens": 39911564.0,
      "step": 19225
    },
    {
      "entropy": 5.763419914245605,
      "epoch": 1.7443759071117562,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004697780140283848,
      "loss": 5.3909,
      "mean_token_accuracy": 0.1672835573554039,
      "num_tokens": 39922168.0,
      "step": 19230
    },
    {
      "entropy": 5.754757404327393,
      "epoch": 1.7448294629898404,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004697618144659683,
      "loss": 5.4359,
      "mean_token_accuracy": 0.17228929251432418,
      "num_tokens": 39932690.0,
      "step": 19235
    },
    {
      "entropy": 5.826336765289307,
      "epoch": 1.7452830188679245,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004697456108758278,
      "loss": 5.6122,
      "mean_token_accuracy": 0.16233527511358262,
      "num_tokens": 39942970.0,
      "step": 19240
    },
    {
      "entropy": 5.727045917510987,
      "epoch": 1.7457365747460087,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004697294032582986,
      "loss": 5.4397,
      "mean_token_accuracy": 0.16590743958950044,
      "num_tokens": 39953255.0,
      "step": 19245
    },
    {
      "entropy": 5.7550865650177006,
      "epoch": 1.7461901306240928,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046971319161371564,
      "loss": 5.3851,
      "mean_token_accuracy": 0.17084863036870956,
      "num_tokens": 39963521.0,
      "step": 19250
    },
    {
      "entropy": 5.775255632400513,
      "epoch": 1.746643686502177,
      "grad_norm": 16.5,
      "learning_rate": 0.00046969697594241426,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16296930462121964,
      "num_tokens": 39973494.0,
      "step": 19255
    },
    {
      "entropy": 5.768852090835571,
      "epoch": 1.7470972423802613,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004696807562447298,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1571136325597763,
      "num_tokens": 39984652.0,
      "step": 19260
    },
    {
      "entropy": 5.825593900680542,
      "epoch": 1.7475507982583456,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046966453252099763,
      "loss": 5.5307,
      "mean_token_accuracy": 0.1662019520998001,
      "num_tokens": 39995490.0,
      "step": 19265
    },
    {
      "entropy": 5.749036455154419,
      "epoch": 1.7480043541364296,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046964830477155335,
      "loss": 5.3248,
      "mean_token_accuracy": 0.18369510769844055,
      "num_tokens": 40006447.0,
      "step": 19270
    },
    {
      "entropy": 5.854143238067627,
      "epoch": 1.7484579100145137,
      "grad_norm": 3.1875,
      "learning_rate": 0.00046963207299673257,
      "loss": 5.542,
      "mean_token_accuracy": 0.16635467112064362,
      "num_tokens": 40015733.0,
      "step": 19275
    },
    {
      "entropy": 5.78511643409729,
      "epoch": 1.748911465892598,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046961583719687085,
      "loss": 5.4216,
      "mean_token_accuracy": 0.16645727008581163,
      "num_tokens": 40025844.0,
      "step": 19280
    },
    {
      "entropy": 5.708956527709961,
      "epoch": 1.7493650217706822,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046959959737230414,
      "loss": 5.4691,
      "mean_token_accuracy": 0.17176474183797835,
      "num_tokens": 40035942.0,
      "step": 19285
    },
    {
      "entropy": 5.692073822021484,
      "epoch": 1.7498185776487665,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046958335352336806,
      "loss": 5.3583,
      "mean_token_accuracy": 0.17035921514034272,
      "num_tokens": 40045551.0,
      "step": 19290
    },
    {
      "entropy": 5.773750638961792,
      "epoch": 1.7502721335268505,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004695671056503987,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1748579889535904,
      "num_tokens": 40055803.0,
      "step": 19295
    },
    {
      "entropy": 5.734988975524902,
      "epoch": 1.7507256894049346,
      "grad_norm": 2.796875,
      "learning_rate": 0.000469550853753732,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16294994056224824,
      "num_tokens": 40066004.0,
      "step": 19300
    },
    {
      "entropy": 5.763793134689331,
      "epoch": 1.7511792452830188,
      "grad_norm": 2.96875,
      "learning_rate": 0.000469534597833704,
      "loss": 5.4627,
      "mean_token_accuracy": 0.17278390526771545,
      "num_tokens": 40075522.0,
      "step": 19305
    },
    {
      "entropy": 5.753098106384277,
      "epoch": 1.751632801161103,
      "grad_norm": 3.015625,
      "learning_rate": 0.000469518337890651,
      "loss": 5.4461,
      "mean_token_accuracy": 0.17342288494110109,
      "num_tokens": 40084839.0,
      "step": 19310
    },
    {
      "entropy": 5.746442699432373,
      "epoch": 1.7520863570391874,
      "grad_norm": 4.9375,
      "learning_rate": 0.00046950207392490917,
      "loss": 5.4803,
      "mean_token_accuracy": 0.167057403922081,
      "num_tokens": 40094392.0,
      "step": 19315
    },
    {
      "entropy": 5.705888652801514,
      "epoch": 1.7525399129172714,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046948580593681496,
      "loss": 5.4134,
      "mean_token_accuracy": 0.17006176263093947,
      "num_tokens": 40105990.0,
      "step": 19320
    },
    {
      "entropy": 5.635200881958008,
      "epoch": 1.7529934687953554,
      "grad_norm": 2.625,
      "learning_rate": 0.0004694695339267047,
      "loss": 5.3652,
      "mean_token_accuracy": 0.1706249088048935,
      "num_tokens": 40116143.0,
      "step": 19325
    },
    {
      "entropy": 5.837895059585572,
      "epoch": 1.7534470246734397,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046945325789491483,
      "loss": 5.5377,
      "mean_token_accuracy": 0.16725095808506013,
      "num_tokens": 40126849.0,
      "step": 19330
    },
    {
      "entropy": 5.809238910675049,
      "epoch": 1.753900580551524,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004694369778417822,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1618003949522972,
      "num_tokens": 40137259.0,
      "step": 19335
    },
    {
      "entropy": 5.6992833614349365,
      "epoch": 1.7543541364296082,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004694206937676431,
      "loss": 5.3519,
      "mean_token_accuracy": 0.16559149771928788,
      "num_tokens": 40148578.0,
      "step": 19340
    },
    {
      "entropy": 5.767865705490112,
      "epoch": 1.7548076923076923,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004694044056728347,
      "loss": 5.5024,
      "mean_token_accuracy": 0.15882248431444168,
      "num_tokens": 40159754.0,
      "step": 19345
    },
    {
      "entropy": 5.793983793258667,
      "epoch": 1.7552612481857763,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046938811355769363,
      "loss": 5.4838,
      "mean_token_accuracy": 0.1664532572031021,
      "num_tokens": 40169388.0,
      "step": 19350
    },
    {
      "entropy": 5.7318675994873045,
      "epoch": 1.7557148040638606,
      "grad_norm": 3.453125,
      "learning_rate": 0.00046937181742255673,
      "loss": 5.4717,
      "mean_token_accuracy": 0.16685234010219574,
      "num_tokens": 40179271.0,
      "step": 19355
    },
    {
      "entropy": 5.744947814941407,
      "epoch": 1.7561683599419449,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004693555172677612,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16356551945209502,
      "num_tokens": 40189292.0,
      "step": 19360
    },
    {
      "entropy": 5.718231630325318,
      "epoch": 1.7566219158200291,
      "grad_norm": 1.53125,
      "learning_rate": 0.000469339213093644,
      "loss": 5.4681,
      "mean_token_accuracy": 0.17023038417100905,
      "num_tokens": 40199271.0,
      "step": 19365
    },
    {
      "entropy": 5.758769369125366,
      "epoch": 1.7570754716981132,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046932290490054235,
      "loss": 5.5027,
      "mean_token_accuracy": 0.16500424295663835,
      "num_tokens": 40210386.0,
      "step": 19370
    },
    {
      "entropy": 5.853041124343872,
      "epoch": 1.7575290275761972,
      "grad_norm": 3.25,
      "learning_rate": 0.0004693065926887936,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1652991831302643,
      "num_tokens": 40220892.0,
      "step": 19375
    },
    {
      "entropy": 5.7732137680053714,
      "epoch": 1.7579825834542815,
      "grad_norm": 2.0,
      "learning_rate": 0.00046929027645873484,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16496799886226654,
      "num_tokens": 40231737.0,
      "step": 19380
    },
    {
      "entropy": 5.697567462921143,
      "epoch": 1.7584361393323658,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004692739562107038,
      "loss": 5.4959,
      "mean_token_accuracy": 0.16431961506605147,
      "num_tokens": 40242218.0,
      "step": 19385
    },
    {
      "entropy": 5.76456651687622,
      "epoch": 1.75888969521045,
      "grad_norm": 1.875,
      "learning_rate": 0.00046925763194503775,
      "loss": 5.4942,
      "mean_token_accuracy": 0.16576229482889177,
      "num_tokens": 40253659.0,
      "step": 19390
    },
    {
      "entropy": 5.746914958953857,
      "epoch": 1.759343251088534,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004692413036620744,
      "loss": 5.3469,
      "mean_token_accuracy": 0.17357626110315322,
      "num_tokens": 40264328.0,
      "step": 19395
    },
    {
      "entropy": 5.629584264755249,
      "epoch": 1.7597968069666183,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004692249713621514,
      "loss": 5.4403,
      "mean_token_accuracy": 0.16970434933900833,
      "num_tokens": 40275188.0,
      "step": 19400
    },
    {
      "entropy": 5.740295553207398,
      "epoch": 1.7602503628447024,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046920863504560647,
      "loss": 5.6045,
      "mean_token_accuracy": 0.159642094373703,
      "num_tokens": 40287094.0,
      "step": 19405
    },
    {
      "entropy": 5.823371458053589,
      "epoch": 1.7607039187227866,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004691922947127775,
      "loss": 5.4397,
      "mean_token_accuracy": 0.16732596904039382,
      "num_tokens": 40296551.0,
      "step": 19410
    },
    {
      "entropy": 5.746456289291382,
      "epoch": 1.761157474600871,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004691759503640024,
      "loss": 5.4454,
      "mean_token_accuracy": 0.16325863152742387,
      "num_tokens": 40306920.0,
      "step": 19415
    },
    {
      "entropy": 5.70514326095581,
      "epoch": 1.761611030478955,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004691596019996191,
      "loss": 5.3486,
      "mean_token_accuracy": 0.17629866451025009,
      "num_tokens": 40316797.0,
      "step": 19420
    },
    {
      "entropy": 5.653779125213623,
      "epoch": 1.7620645863570392,
      "grad_norm": 2.375,
      "learning_rate": 0.0004691432496199658,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1668100506067276,
      "num_tokens": 40327002.0,
      "step": 19425
    },
    {
      "entropy": 5.850853490829468,
      "epoch": 1.7625181422351233,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046912689322538064,
      "loss": 5.5736,
      "mean_token_accuracy": 0.16288373619318008,
      "num_tokens": 40337247.0,
      "step": 19430
    },
    {
      "entropy": 5.812981128692627,
      "epoch": 1.7629716981132075,
      "grad_norm": 2.625,
      "learning_rate": 0.00046911053281620187,
      "loss": 5.5192,
      "mean_token_accuracy": 0.16758355349302292,
      "num_tokens": 40346960.0,
      "step": 19435
    },
    {
      "entropy": 5.786733675003052,
      "epoch": 1.7634252539912918,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046909416839276775,
      "loss": 5.5074,
      "mean_token_accuracy": 0.16488222628831864,
      "num_tokens": 40357169.0,
      "step": 19440
    },
    {
      "entropy": 5.654955434799194,
      "epoch": 1.763878809869376,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004690777999554168,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1766969934105873,
      "num_tokens": 40367660.0,
      "step": 19445
    },
    {
      "entropy": 5.715146589279175,
      "epoch": 1.76433236574746,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004690614275044875,
      "loss": 5.3636,
      "mean_token_accuracy": 0.1688097044825554,
      "num_tokens": 40378217.0,
      "step": 19450
    },
    {
      "entropy": 5.717126512527466,
      "epoch": 1.7647859216255442,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004690450510403184,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1677378997206688,
      "num_tokens": 40388249.0,
      "step": 19455
    },
    {
      "entropy": 5.832857275009156,
      "epoch": 1.7652394775036284,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004690286705632483,
      "loss": 5.5606,
      "mean_token_accuracy": 0.16159394830465318,
      "num_tokens": 40398239.0,
      "step": 19460
    },
    {
      "entropy": 5.7639219760894775,
      "epoch": 1.7656930333817127,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004690122860736157,
      "loss": 5.5252,
      "mean_token_accuracy": 0.16180956810712815,
      "num_tokens": 40408473.0,
      "step": 19465
    },
    {
      "entropy": 5.814750385284424,
      "epoch": 1.766146589259797,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046899589757175964,
      "loss": 5.5689,
      "mean_token_accuracy": 0.16638534814119338,
      "num_tokens": 40418674.0,
      "step": 19470
    },
    {
      "entropy": 5.718127632141114,
      "epoch": 1.766600145137881,
      "grad_norm": 1.921875,
      "learning_rate": 0.000468979505058019,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16941707581281662,
      "num_tokens": 40429815.0,
      "step": 19475
    },
    {
      "entropy": 5.814056348800659,
      "epoch": 1.767053701015965,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046896310853273274,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16955206245183946,
      "num_tokens": 40439809.0,
      "step": 19480
    },
    {
      "entropy": 5.852337598800659,
      "epoch": 1.7675072568940493,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046894670799624,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1665671154856682,
      "num_tokens": 40449503.0,
      "step": 19485
    },
    {
      "entropy": 5.703304052352905,
      "epoch": 1.7679608127721336,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046893030344887995,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1586090013384819,
      "num_tokens": 40459794.0,
      "step": 19490
    },
    {
      "entropy": 5.74628643989563,
      "epoch": 1.7684143686502178,
      "grad_norm": 3.109375,
      "learning_rate": 0.00046891389489099166,
      "loss": 5.4367,
      "mean_token_accuracy": 0.17245279997587204,
      "num_tokens": 40470044.0,
      "step": 19495
    },
    {
      "entropy": 5.777000808715821,
      "epoch": 1.7688679245283019,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046889748232291476,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1666676953434944,
      "num_tokens": 40480709.0,
      "step": 19500
    },
    {
      "entropy": 5.639666128158569,
      "epoch": 1.769321480406386,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046888106574498843,
      "loss": 5.2925,
      "mean_token_accuracy": 0.17601066827774048,
      "num_tokens": 40490906.0,
      "step": 19505
    },
    {
      "entropy": 5.665496158599853,
      "epoch": 1.7697750362844702,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004688646451575522,
      "loss": 5.3903,
      "mean_token_accuracy": 0.17651761025190355,
      "num_tokens": 40501971.0,
      "step": 19510
    },
    {
      "entropy": 5.729297351837158,
      "epoch": 1.7702285921625545,
      "grad_norm": 4.65625,
      "learning_rate": 0.00046884822056094587,
      "loss": 5.5064,
      "mean_token_accuracy": 0.16194332540035247,
      "num_tokens": 40512685.0,
      "step": 19515
    },
    {
      "entropy": 5.685367727279663,
      "epoch": 1.7706821480406387,
      "grad_norm": 3.421875,
      "learning_rate": 0.00046883179195550883,
      "loss": 5.4225,
      "mean_token_accuracy": 0.17297301590442657,
      "num_tokens": 40523460.0,
      "step": 19520
    },
    {
      "entropy": 5.753216361999511,
      "epoch": 1.7711357039187228,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046881535934158094,
      "loss": 5.4108,
      "mean_token_accuracy": 0.17228614389896393,
      "num_tokens": 40533594.0,
      "step": 19525
    },
    {
      "entropy": 5.716214323043824,
      "epoch": 1.7715892597968068,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004687989227195021,
      "loss": 5.4686,
      "mean_token_accuracy": 0.16805912554264069,
      "num_tokens": 40543504.0,
      "step": 19530
    },
    {
      "entropy": 5.7357550144195555,
      "epoch": 1.772042815674891,
      "grad_norm": 2.5,
      "learning_rate": 0.00046878248208961204,
      "loss": 5.4105,
      "mean_token_accuracy": 0.1694680079817772,
      "num_tokens": 40553711.0,
      "step": 19535
    },
    {
      "entropy": 5.755251741409301,
      "epoch": 1.7724963715529753,
      "grad_norm": 3.09375,
      "learning_rate": 0.000468766037452251,
      "loss": 5.4723,
      "mean_token_accuracy": 0.16672378033399582,
      "num_tokens": 40563731.0,
      "step": 19540
    },
    {
      "entropy": 5.748689413070679,
      "epoch": 1.7729499274310596,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046874958880775886,
      "loss": 5.4889,
      "mean_token_accuracy": 0.1663608193397522,
      "num_tokens": 40574235.0,
      "step": 19545
    },
    {
      "entropy": 5.710897970199585,
      "epoch": 1.7734034833091437,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004687331361564758,
      "loss": 5.3908,
      "mean_token_accuracy": 0.17288807779550552,
      "num_tokens": 40584459.0,
      "step": 19550
    },
    {
      "entropy": 5.690310621261597,
      "epoch": 1.7738570391872277,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004687166794987422,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1673416405916214,
      "num_tokens": 40594268.0,
      "step": 19555
    },
    {
      "entropy": 5.72938232421875,
      "epoch": 1.774310595065312,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004687002188348983,
      "loss": 5.4334,
      "mean_token_accuracy": 0.17143257111310958,
      "num_tokens": 40604673.0,
      "step": 19560
    },
    {
      "entropy": 5.6319482803344725,
      "epoch": 1.7747641509433962,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046868375416528443,
      "loss": 5.4061,
      "mean_token_accuracy": 0.16587171256542205,
      "num_tokens": 40614998.0,
      "step": 19565
    },
    {
      "entropy": 5.724674463272095,
      "epoch": 1.7752177068214805,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046866728549024123,
      "loss": 5.4493,
      "mean_token_accuracy": 0.16891325563192366,
      "num_tokens": 40624729.0,
      "step": 19570
    },
    {
      "entropy": 5.778536176681518,
      "epoch": 1.7756712626995645,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004686508128101092,
      "loss": 5.5457,
      "mean_token_accuracy": 0.16445342302322388,
      "num_tokens": 40635673.0,
      "step": 19575
    },
    {
      "entropy": 5.786128377914428,
      "epoch": 1.7761248185776488,
      "grad_norm": 3.53125,
      "learning_rate": 0.000468634336125229,
      "loss": 5.5042,
      "mean_token_accuracy": 0.1596837192773819,
      "num_tokens": 40645990.0,
      "step": 19580
    },
    {
      "entropy": 5.712848901748657,
      "epoch": 1.7765783744557329,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046861785543594143,
      "loss": 5.4685,
      "mean_token_accuracy": 0.17290018498897552,
      "num_tokens": 40656339.0,
      "step": 19585
    },
    {
      "entropy": 5.7400017261505125,
      "epoch": 1.7770319303338171,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004686013707425873,
      "loss": 5.3573,
      "mean_token_accuracy": 0.1747355192899704,
      "num_tokens": 40666371.0,
      "step": 19590
    },
    {
      "entropy": 5.738454675674438,
      "epoch": 1.7774854862119014,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046858488204550735,
      "loss": 5.422,
      "mean_token_accuracy": 0.1706037238240242,
      "num_tokens": 40676423.0,
      "step": 19595
    },
    {
      "entropy": 5.696428346633911,
      "epoch": 1.7779390420899854,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004685683893450428,
      "loss": 5.3974,
      "mean_token_accuracy": 0.1714494213461876,
      "num_tokens": 40686331.0,
      "step": 19600
    },
    {
      "entropy": 5.648662757873535,
      "epoch": 1.7783925979680697,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004685518926415346,
      "loss": 5.435,
      "mean_token_accuracy": 0.17380552589893342,
      "num_tokens": 40697408.0,
      "step": 19605
    },
    {
      "entropy": 5.637779664993286,
      "epoch": 1.7788461538461537,
      "grad_norm": 2.21875,
      "learning_rate": 0.000468535391935324,
      "loss": 5.334,
      "mean_token_accuracy": 0.1766116961836815,
      "num_tokens": 40707078.0,
      "step": 19610
    },
    {
      "entropy": 5.770530891418457,
      "epoch": 1.779299709724238,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004685188872267521,
      "loss": 5.459,
      "mean_token_accuracy": 0.17023564428091048,
      "num_tokens": 40718116.0,
      "step": 19615
    },
    {
      "entropy": 5.7449322700500485,
      "epoch": 1.7797532656023223,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004685023785161603,
      "loss": 5.4542,
      "mean_token_accuracy": 0.16756798774003984,
      "num_tokens": 40727737.0,
      "step": 19620
    },
    {
      "entropy": 5.706567668914795,
      "epoch": 1.7802068214804065,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004684858658038901,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1664900004863739,
      "num_tokens": 40737455.0,
      "step": 19625
    },
    {
      "entropy": 5.763713693618774,
      "epoch": 1.7806603773584906,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004684693490902828,
      "loss": 5.4525,
      "mean_token_accuracy": 0.17001446783542634,
      "num_tokens": 40746715.0,
      "step": 19630
    },
    {
      "entropy": 5.620311737060547,
      "epoch": 1.7811139332365746,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046845282837568,
      "loss": 5.326,
      "mean_token_accuracy": 0.173885278403759,
      "num_tokens": 40756432.0,
      "step": 19635
    },
    {
      "entropy": 5.601522159576416,
      "epoch": 1.781567489114659,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046843630366042355,
      "loss": 5.4179,
      "mean_token_accuracy": 0.16922337859869002,
      "num_tokens": 40766629.0,
      "step": 19640
    },
    {
      "entropy": 5.85070309638977,
      "epoch": 1.7820210449927432,
      "grad_norm": 4.75,
      "learning_rate": 0.00046841977494485506,
      "loss": 5.5397,
      "mean_token_accuracy": 0.1639332041144371,
      "num_tokens": 40776650.0,
      "step": 19645
    },
    {
      "entropy": 5.8477452278137205,
      "epoch": 1.7824746008708274,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004684032422293162,
      "loss": 5.4272,
      "mean_token_accuracy": 0.17664769738912584,
      "num_tokens": 40786722.0,
      "step": 19650
    },
    {
      "entropy": 5.632713794708252,
      "epoch": 1.7829281567489115,
      "grad_norm": 3.3125,
      "learning_rate": 0.000468386705514149,
      "loss": 5.2893,
      "mean_token_accuracy": 0.17964501231908797,
      "num_tokens": 40797058.0,
      "step": 19655
    },
    {
      "entropy": 5.678776836395263,
      "epoch": 1.7833817126269955,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004683701647996956,
      "loss": 5.4914,
      "mean_token_accuracy": 0.1709525018930435,
      "num_tokens": 40808207.0,
      "step": 19660
    },
    {
      "entropy": 5.8047943115234375,
      "epoch": 1.7838352685050798,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004683536200862978,
      "loss": 5.5425,
      "mean_token_accuracy": 0.16240738034248353,
      "num_tokens": 40819655.0,
      "step": 19665
    },
    {
      "entropy": 5.78357515335083,
      "epoch": 1.784288824383164,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004683370713742979,
      "loss": 5.538,
      "mean_token_accuracy": 0.16666995584964753,
      "num_tokens": 40831015.0,
      "step": 19670
    },
    {
      "entropy": 5.744060087203979,
      "epoch": 1.7847423802612483,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046832051866403814,
      "loss": 5.4183,
      "mean_token_accuracy": 0.16959428340196608,
      "num_tokens": 40841763.0,
      "step": 19675
    },
    {
      "entropy": 5.720825290679931,
      "epoch": 1.7851959361393324,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004683039619558607,
      "loss": 5.4221,
      "mean_token_accuracy": 0.16717845052480698,
      "num_tokens": 40852479.0,
      "step": 19680
    },
    {
      "entropy": 5.759423112869262,
      "epoch": 1.7856494920174164,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046828740125010813,
      "loss": 5.4476,
      "mean_token_accuracy": 0.17145468443632125,
      "num_tokens": 40862486.0,
      "step": 19685
    },
    {
      "entropy": 5.776780509948731,
      "epoch": 1.7861030478955007,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004682708365471228,
      "loss": 5.538,
      "mean_token_accuracy": 0.16580802202224731,
      "num_tokens": 40872182.0,
      "step": 19690
    },
    {
      "entropy": 5.793340063095092,
      "epoch": 1.786556603773585,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046825426784724734,
      "loss": 5.5092,
      "mean_token_accuracy": 0.1669277548789978,
      "num_tokens": 40883036.0,
      "step": 19695
    },
    {
      "entropy": 5.784368181228638,
      "epoch": 1.7870101596516692,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004682376951508244,
      "loss": 5.5278,
      "mean_token_accuracy": 0.15579292476177214,
      "num_tokens": 40893684.0,
      "step": 19700
    },
    {
      "entropy": 5.810639333724976,
      "epoch": 1.7874637155297532,
      "grad_norm": 2.5,
      "learning_rate": 0.00046822111845819665,
      "loss": 5.5253,
      "mean_token_accuracy": 0.16733533889055252,
      "num_tokens": 40904540.0,
      "step": 19705
    },
    {
      "entropy": 5.792911243438721,
      "epoch": 1.7879172714078373,
      "grad_norm": 1.75,
      "learning_rate": 0.00046820453776970697,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16589679569005966,
      "num_tokens": 40915269.0,
      "step": 19710
    },
    {
      "entropy": 5.7617835521698,
      "epoch": 1.7883708272859216,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046818795308569815,
      "loss": 5.5002,
      "mean_token_accuracy": 0.16172169893980026,
      "num_tokens": 40925987.0,
      "step": 19715
    },
    {
      "entropy": 5.721894836425781,
      "epoch": 1.7888243831640058,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004681713644065133,
      "loss": 5.4157,
      "mean_token_accuracy": 0.1773853600025177,
      "num_tokens": 40936130.0,
      "step": 19720
    },
    {
      "entropy": 5.641699981689453,
      "epoch": 1.78927793904209,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004681547717324954,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1718267723917961,
      "num_tokens": 40945476.0,
      "step": 19725
    },
    {
      "entropy": 5.607181644439697,
      "epoch": 1.7897314949201741,
      "grad_norm": 2.984375,
      "learning_rate": 0.00046813817506398755,
      "loss": 5.3425,
      "mean_token_accuracy": 0.17605352252721787,
      "num_tokens": 40956492.0,
      "step": 19730
    },
    {
      "entropy": 5.79252290725708,
      "epoch": 1.7901850507982582,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046812157440133303,
      "loss": 5.5327,
      "mean_token_accuracy": 0.1630358874797821,
      "num_tokens": 40966990.0,
      "step": 19735
    },
    {
      "entropy": 5.789464998245239,
      "epoch": 1.7906386066763424,
      "grad_norm": 4.25,
      "learning_rate": 0.0004681049697448752,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16948862820863725,
      "num_tokens": 40977065.0,
      "step": 19740
    },
    {
      "entropy": 5.719782161712646,
      "epoch": 1.7910921625544267,
      "grad_norm": 3.15625,
      "learning_rate": 0.00046808836109495734,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1673167437314987,
      "num_tokens": 40987207.0,
      "step": 19745
    },
    {
      "entropy": 5.727013969421387,
      "epoch": 1.791545718432511,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004680717484519229,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1701340988278389,
      "num_tokens": 40997541.0,
      "step": 19750
    },
    {
      "entropy": 5.728847789764404,
      "epoch": 1.791999274310595,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004680551318161156,
      "loss": 5.4507,
      "mean_token_accuracy": 0.17273229211568833,
      "num_tokens": 41009044.0,
      "step": 19755
    },
    {
      "entropy": 5.71229887008667,
      "epoch": 1.7924528301886793,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046803851118787897,
      "loss": 5.4263,
      "mean_token_accuracy": 0.17489436715841294,
      "num_tokens": 41018165.0,
      "step": 19760
    },
    {
      "entropy": 5.677169752120972,
      "epoch": 1.7929063860667633,
      "grad_norm": 7.125,
      "learning_rate": 0.0004680218865675567,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1694793716073036,
      "num_tokens": 41028972.0,
      "step": 19765
    },
    {
      "entropy": 5.719891786575317,
      "epoch": 1.7933599419448476,
      "grad_norm": 2.75,
      "learning_rate": 0.0004680052579554926,
      "loss": 5.4402,
      "mean_token_accuracy": 0.16840966939926147,
      "num_tokens": 41039552.0,
      "step": 19770
    },
    {
      "entropy": 5.806222152709961,
      "epoch": 1.7938134978229319,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004679886253520307,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1612916558980942,
      "num_tokens": 41049808.0,
      "step": 19775
    },
    {
      "entropy": 5.755672264099121,
      "epoch": 1.794267053701016,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046797198875751474,
      "loss": 5.4035,
      "mean_token_accuracy": 0.1624643474817276,
      "num_tokens": 41060574.0,
      "step": 19780
    },
    {
      "entropy": 5.797921133041382,
      "epoch": 1.7947206095791002,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004679553481722889,
      "loss": 5.5656,
      "mean_token_accuracy": 0.16127157658338548,
      "num_tokens": 41071239.0,
      "step": 19785
    },
    {
      "entropy": 5.762017679214478,
      "epoch": 1.7951741654571842,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046793870359669733,
      "loss": 5.5592,
      "mean_token_accuracy": 0.16535107791423798,
      "num_tokens": 41080819.0,
      "step": 19790
    },
    {
      "entropy": 5.80654878616333,
      "epoch": 1.7956277213352685,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004679220550310842,
      "loss": 5.3951,
      "mean_token_accuracy": 0.17594811469316482,
      "num_tokens": 41090457.0,
      "step": 19795
    },
    {
      "entropy": 5.762312173843384,
      "epoch": 1.7960812772133528,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004679054024757937,
      "loss": 5.4406,
      "mean_token_accuracy": 0.17273680865764618,
      "num_tokens": 41101210.0,
      "step": 19800
    },
    {
      "entropy": 5.743722724914551,
      "epoch": 1.796534833091437,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004678887459311704,
      "loss": 5.4685,
      "mean_token_accuracy": 0.17225315570831298,
      "num_tokens": 41110566.0,
      "step": 19805
    },
    {
      "entropy": 5.710873460769653,
      "epoch": 1.796988388969521,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004678720853975587,
      "loss": 5.3733,
      "mean_token_accuracy": 0.17360036373138427,
      "num_tokens": 41120000.0,
      "step": 19810
    },
    {
      "entropy": 5.680813884735107,
      "epoch": 1.797441944847605,
      "grad_norm": 5.40625,
      "learning_rate": 0.00046785542087530307,
      "loss": 5.3451,
      "mean_token_accuracy": 0.17854238003492356,
      "num_tokens": 41129944.0,
      "step": 19815
    },
    {
      "entropy": 5.669245719909668,
      "epoch": 1.7978955007256894,
      "grad_norm": 2.125,
      "learning_rate": 0.0004678387523647483,
      "loss": 5.4284,
      "mean_token_accuracy": 0.162130106985569,
      "num_tokens": 41138803.0,
      "step": 19820
    },
    {
      "entropy": 5.745019483566284,
      "epoch": 1.7983490566037736,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046782207986623885,
      "loss": 5.5222,
      "mean_token_accuracy": 0.15802081674337387,
      "num_tokens": 41150365.0,
      "step": 19825
    },
    {
      "entropy": 5.839174079895019,
      "epoch": 1.798802612481858,
      "grad_norm": 6.0,
      "learning_rate": 0.00046780540338011974,
      "loss": 5.4774,
      "mean_token_accuracy": 0.16420210599899293,
      "num_tokens": 41160310.0,
      "step": 19830
    },
    {
      "entropy": 5.73025484085083,
      "epoch": 1.799256168359942,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004677887229067358,
      "loss": 5.3485,
      "mean_token_accuracy": 0.17253633141517638,
      "num_tokens": 41170506.0,
      "step": 19835
    },
    {
      "entropy": 5.748581743240356,
      "epoch": 1.799709724238026,
      "grad_norm": 3.796875,
      "learning_rate": 0.00046777203844643184,
      "loss": 5.4687,
      "mean_token_accuracy": 0.16226973682641982,
      "num_tokens": 41181293.0,
      "step": 19840
    },
    {
      "entropy": 5.6895441055297855,
      "epoch": 1.8001632801161103,
      "grad_norm": 2.921875,
      "learning_rate": 0.00046775534999955303,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16690000146627426,
      "num_tokens": 41191637.0,
      "step": 19845
    },
    {
      "entropy": 5.670256042480469,
      "epoch": 1.8006168359941945,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004677386575664445,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1756252184510231,
      "num_tokens": 41201833.0,
      "step": 19850
    },
    {
      "entropy": 5.800416707992554,
      "epoch": 1.8010703918722788,
      "grad_norm": 2.96875,
      "learning_rate": 0.00046772196114745133,
      "loss": 5.4677,
      "mean_token_accuracy": 0.17310334891080856,
      "num_tokens": 41211243.0,
      "step": 19855
    },
    {
      "entropy": 5.778508853912354,
      "epoch": 1.8015239477503628,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004677052607429189,
      "loss": 5.4416,
      "mean_token_accuracy": 0.17324743568897247,
      "num_tokens": 41221111.0,
      "step": 19860
    },
    {
      "entropy": 5.643940353393555,
      "epoch": 1.8019775036284469,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004676885563531927,
      "loss": 5.4421,
      "mean_token_accuracy": 0.1686961233615875,
      "num_tokens": 41231578.0,
      "step": 19865
    },
    {
      "entropy": 5.693081045150757,
      "epoch": 1.8024310595065312,
      "grad_norm": 6.59375,
      "learning_rate": 0.00046767184797861795,
      "loss": 5.504,
      "mean_token_accuracy": 0.17197534590959548,
      "num_tokens": 41242130.0,
      "step": 19870
    },
    {
      "entropy": 5.801265478134155,
      "epoch": 1.8028846153846154,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046765513561954026,
      "loss": 5.4562,
      "mean_token_accuracy": 0.16555435657501222,
      "num_tokens": 41252885.0,
      "step": 19875
    },
    {
      "entropy": 5.800195503234863,
      "epoch": 1.8033381712626997,
      "grad_norm": 3.0625,
      "learning_rate": 0.00046763841927630534,
      "loss": 5.4796,
      "mean_token_accuracy": 0.16538225412368773,
      "num_tokens": 41263532.0,
      "step": 19880
    },
    {
      "entropy": 5.795073509216309,
      "epoch": 1.8037917271407837,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046762169894925876,
      "loss": 5.4533,
      "mean_token_accuracy": 0.16939304918050765,
      "num_tokens": 41273469.0,
      "step": 19885
    },
    {
      "entropy": 5.723706531524658,
      "epoch": 1.8042452830188678,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004676049746387464,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1643285259604454,
      "num_tokens": 41283663.0,
      "step": 19890
    },
    {
      "entropy": 5.689121675491333,
      "epoch": 1.804698838896952,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004675882463451141,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1719444215297699,
      "num_tokens": 41294657.0,
      "step": 19895
    },
    {
      "entropy": 5.796807861328125,
      "epoch": 1.8051523947750363,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004675715140687078,
      "loss": 5.477,
      "mean_token_accuracy": 0.16933890730142592,
      "num_tokens": 41305124.0,
      "step": 19900
    },
    {
      "entropy": 5.685372400283813,
      "epoch": 1.8056059506531206,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046755477780987344,
      "loss": 5.3867,
      "mean_token_accuracy": 0.17308646738529204,
      "num_tokens": 41315839.0,
      "step": 19905
    },
    {
      "entropy": 5.768025636672974,
      "epoch": 1.8060595065312046,
      "grad_norm": 4.625,
      "learning_rate": 0.00046753803756895726,
      "loss": 5.4721,
      "mean_token_accuracy": 0.16663598865270615,
      "num_tokens": 41326388.0,
      "step": 19910
    },
    {
      "entropy": 5.773540735244751,
      "epoch": 1.8065130624092887,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004675212933463054,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1638411819934845,
      "num_tokens": 41337333.0,
      "step": 19915
    },
    {
      "entropy": 5.753086757659912,
      "epoch": 1.806966618287373,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004675045451422641,
      "loss": 5.508,
      "mean_token_accuracy": 0.15967685580253602,
      "num_tokens": 41348353.0,
      "step": 19920
    },
    {
      "entropy": 5.719946622848511,
      "epoch": 1.8074201741654572,
      "grad_norm": 3.5,
      "learning_rate": 0.00046748779295717973,
      "loss": 5.5003,
      "mean_token_accuracy": 0.16829769611358641,
      "num_tokens": 41358797.0,
      "step": 19925
    },
    {
      "entropy": 5.718845510482788,
      "epoch": 1.8078737300435415,
      "grad_norm": 3.125,
      "learning_rate": 0.0004674710367913989,
      "loss": 5.4835,
      "mean_token_accuracy": 0.16793869584798812,
      "num_tokens": 41368818.0,
      "step": 19930
    },
    {
      "entropy": 5.75493540763855,
      "epoch": 1.8083272859216255,
      "grad_norm": 4.0,
      "learning_rate": 0.0004674542766452678,
      "loss": 5.4443,
      "mean_token_accuracy": 0.1690200611948967,
      "num_tokens": 41378949.0,
      "step": 19935
    },
    {
      "entropy": 5.756303310394287,
      "epoch": 1.8087808417997098,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046743751251913327,
      "loss": 5.4155,
      "mean_token_accuracy": 0.16848176568746567,
      "num_tokens": 41388892.0,
      "step": 19940
    },
    {
      "entropy": 5.75516562461853,
      "epoch": 1.8092343976777938,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046742074441334197,
      "loss": 5.4534,
      "mean_token_accuracy": 0.16979209780693055,
      "num_tokens": 41399197.0,
      "step": 19945
    },
    {
      "entropy": 5.7556736946105955,
      "epoch": 1.809687953555878,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004674039723282406,
      "loss": 5.4691,
      "mean_token_accuracy": 0.16462814956903457,
      "num_tokens": 41409551.0,
      "step": 19950
    },
    {
      "entropy": 5.731266641616822,
      "epoch": 1.8101415094339623,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046738719626417607,
      "loss": 5.3711,
      "mean_token_accuracy": 0.16761139035224915,
      "num_tokens": 41418900.0,
      "step": 19955
    },
    {
      "entropy": 5.690343809127808,
      "epoch": 1.8105950653120464,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046737041622149536,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1677675113081932,
      "num_tokens": 41428204.0,
      "step": 19960
    },
    {
      "entropy": 5.695384407043457,
      "epoch": 1.8110486211901307,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004673536322005454,
      "loss": 5.4476,
      "mean_token_accuracy": 0.16417686641216278,
      "num_tokens": 41437922.0,
      "step": 19965
    },
    {
      "entropy": 5.798392629623413,
      "epoch": 1.8115021770682147,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004673368442016732,
      "loss": 5.4452,
      "mean_token_accuracy": 0.16127115339040757,
      "num_tokens": 41447411.0,
      "step": 19970
    },
    {
      "entropy": 5.7967325210571286,
      "epoch": 1.811955732946299,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004673200522252261,
      "loss": 5.5563,
      "mean_token_accuracy": 0.16601260602474213,
      "num_tokens": 41458756.0,
      "step": 19975
    },
    {
      "entropy": 5.717336988449096,
      "epoch": 1.8124092888243832,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004673032562715514,
      "loss": 5.4437,
      "mean_token_accuracy": 0.1684561550617218,
      "num_tokens": 41468610.0,
      "step": 19980
    },
    {
      "entropy": 5.773146486282348,
      "epoch": 1.8128628447024675,
      "grad_norm": 3.0,
      "learning_rate": 0.00046728645634099627,
      "loss": 5.5827,
      "mean_token_accuracy": 0.16089637577533722,
      "num_tokens": 41479849.0,
      "step": 19985
    },
    {
      "entropy": 5.743735980987549,
      "epoch": 1.8133164005805515,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046726965243390825,
      "loss": 5.4704,
      "mean_token_accuracy": 0.16191229373216628,
      "num_tokens": 41490945.0,
      "step": 19990
    },
    {
      "entropy": 5.686665201187134,
      "epoch": 1.8137699564586356,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004672528445506348,
      "loss": 5.3895,
      "mean_token_accuracy": 0.16374073773622513,
      "num_tokens": 41501479.0,
      "step": 19995
    },
    {
      "entropy": 5.746306419372559,
      "epoch": 1.8142235123367199,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004672360326915236,
      "loss": 5.3871,
      "mean_token_accuracy": 0.17671216130256653,
      "num_tokens": 41512447.0,
      "step": 20000
    },
    {
      "entropy": 5.658472919464112,
      "epoch": 1.8146770682148041,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004672192168569222,
      "loss": 5.3632,
      "mean_token_accuracy": 0.1784674495458603,
      "num_tokens": 41521905.0,
      "step": 20005
    },
    {
      "entropy": 5.704546737670898,
      "epoch": 1.8151306240928884,
      "grad_norm": 4.8125,
      "learning_rate": 0.00046720239704717845,
      "loss": 5.4609,
      "mean_token_accuracy": 0.17121892124414445,
      "num_tokens": 41533143.0,
      "step": 20010
    },
    {
      "entropy": 5.735324287414551,
      "epoch": 1.8155841799709724,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046718557326264,
      "loss": 5.4029,
      "mean_token_accuracy": 0.16707442700862885,
      "num_tokens": 41542234.0,
      "step": 20015
    },
    {
      "entropy": 5.795846652984619,
      "epoch": 1.8160377358490565,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004671687455036551,
      "loss": 5.5362,
      "mean_token_accuracy": 0.16146889850497245,
      "num_tokens": 41552632.0,
      "step": 20020
    },
    {
      "entropy": 5.743000841140747,
      "epoch": 1.8164912917271407,
      "grad_norm": 2.75,
      "learning_rate": 0.0004671519137705715,
      "loss": 5.4546,
      "mean_token_accuracy": 0.17180010229349135,
      "num_tokens": 41563723.0,
      "step": 20025
    },
    {
      "entropy": 5.718971490859985,
      "epoch": 1.816944847605225,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046713507806373736,
      "loss": 5.431,
      "mean_token_accuracy": 0.16718136966228486,
      "num_tokens": 41573680.0,
      "step": 20030
    },
    {
      "entropy": 5.716943073272705,
      "epoch": 1.8173984034833093,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046711823838350086,
      "loss": 5.4921,
      "mean_token_accuracy": 0.16723393350839616,
      "num_tokens": 41584722.0,
      "step": 20035
    },
    {
      "entropy": 5.782856559753418,
      "epoch": 1.8178519593613933,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046710139473021025,
      "loss": 5.496,
      "mean_token_accuracy": 0.1620820641517639,
      "num_tokens": 41595785.0,
      "step": 20040
    },
    {
      "entropy": 5.725548934936524,
      "epoch": 1.8183055152394774,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046708454710421376,
      "loss": 5.52,
      "mean_token_accuracy": 0.16430538892745972,
      "num_tokens": 41606023.0,
      "step": 20045
    },
    {
      "entropy": 5.712940979003906,
      "epoch": 1.8187590711175616,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046706769550586,
      "loss": 5.4393,
      "mean_token_accuracy": 0.16602115631103515,
      "num_tokens": 41616010.0,
      "step": 20050
    },
    {
      "entropy": 5.7980568408966064,
      "epoch": 1.819212626995646,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046705083993549723,
      "loss": 5.401,
      "mean_token_accuracy": 0.1741165891289711,
      "num_tokens": 41625105.0,
      "step": 20055
    },
    {
      "entropy": 5.729220294952393,
      "epoch": 1.8196661828737302,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004670339803934743,
      "loss": 5.455,
      "mean_token_accuracy": 0.1728858545422554,
      "num_tokens": 41634989.0,
      "step": 20060
    },
    {
      "entropy": 5.67266149520874,
      "epoch": 1.8201197387518142,
      "grad_norm": 2.125,
      "learning_rate": 0.0004670171168801396,
      "loss": 5.4369,
      "mean_token_accuracy": 0.16618081033229828,
      "num_tokens": 41645736.0,
      "step": 20065
    },
    {
      "entropy": 5.81215763092041,
      "epoch": 1.8205732946298983,
      "grad_norm": 2.25,
      "learning_rate": 0.000467000249395842,
      "loss": 5.4573,
      "mean_token_accuracy": 0.1677711233496666,
      "num_tokens": 41655505.0,
      "step": 20070
    },
    {
      "entropy": 5.723100280761718,
      "epoch": 1.8210268505079825,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004669833779409303,
      "loss": 5.4128,
      "mean_token_accuracy": 0.17171862870454788,
      "num_tokens": 41665732.0,
      "step": 20075
    },
    {
      "entropy": 5.658949899673462,
      "epoch": 1.8214804063860668,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004669665025157535,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1658420205116272,
      "num_tokens": 41676050.0,
      "step": 20080
    },
    {
      "entropy": 5.796609926223755,
      "epoch": 1.821933962264151,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046694962312066046,
      "loss": 5.5695,
      "mean_token_accuracy": 0.1668984293937683,
      "num_tokens": 41686787.0,
      "step": 20085
    },
    {
      "entropy": 5.721535539627075,
      "epoch": 1.822387518142235,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046693273975600036,
      "loss": 5.3758,
      "mean_token_accuracy": 0.172797729074955,
      "num_tokens": 41697168.0,
      "step": 20090
    },
    {
      "entropy": 5.70417160987854,
      "epoch": 1.8228410740203191,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004669158524221222,
      "loss": 5.4571,
      "mean_token_accuracy": 0.17054543048143386,
      "num_tokens": 41707524.0,
      "step": 20095
    },
    {
      "entropy": 5.762351131439209,
      "epoch": 1.8232946298984034,
      "grad_norm": 3.3125,
      "learning_rate": 0.00046689896111937537,
      "loss": 5.4986,
      "mean_token_accuracy": 0.17491055727005006,
      "num_tokens": 41718109.0,
      "step": 20100
    },
    {
      "entropy": 5.809721040725708,
      "epoch": 1.8237481857764877,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004668820658481091,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1673882022500038,
      "num_tokens": 41728862.0,
      "step": 20105
    },
    {
      "entropy": 5.7927131175994875,
      "epoch": 1.824201741654572,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046686516660867283,
      "loss": 5.545,
      "mean_token_accuracy": 0.16262462735176086,
      "num_tokens": 41739024.0,
      "step": 20110
    },
    {
      "entropy": 5.730351114273072,
      "epoch": 1.824655297532656,
      "grad_norm": 2.875,
      "learning_rate": 0.0004668482634014159,
      "loss": 5.48,
      "mean_token_accuracy": 0.16474743634462358,
      "num_tokens": 41748979.0,
      "step": 20115
    },
    {
      "entropy": 5.750985527038575,
      "epoch": 1.8251088534107403,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004668313562266882,
      "loss": 5.5127,
      "mean_token_accuracy": 0.16831534802913667,
      "num_tokens": 41759972.0,
      "step": 20120
    },
    {
      "entropy": 5.733602046966553,
      "epoch": 1.8255624092888243,
      "grad_norm": 2.1875,
      "learning_rate": 0.000466814445084839,
      "loss": 5.3691,
      "mean_token_accuracy": 0.17213841080665587,
      "num_tokens": 41770150.0,
      "step": 20125
    },
    {
      "entropy": 5.722273159027099,
      "epoch": 1.8260159651669086,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004667975299762183,
      "loss": 5.4044,
      "mean_token_accuracy": 0.17086293399333954,
      "num_tokens": 41780405.0,
      "step": 20130
    },
    {
      "entropy": 5.709363222122192,
      "epoch": 1.8264695210449928,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046678061090117576,
      "loss": 5.3667,
      "mean_token_accuracy": 0.1699816569685936,
      "num_tokens": 41791269.0,
      "step": 20135
    },
    {
      "entropy": 5.654963111877441,
      "epoch": 1.8269230769230769,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004667636878600613,
      "loss": 5.3614,
      "mean_token_accuracy": 0.1758273035287857,
      "num_tokens": 41802513.0,
      "step": 20140
    },
    {
      "entropy": 5.761911296844483,
      "epoch": 1.8273766328011611,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004667467608532249,
      "loss": 5.511,
      "mean_token_accuracy": 0.1597672387957573,
      "num_tokens": 41814149.0,
      "step": 20145
    },
    {
      "entropy": 5.807060623168946,
      "epoch": 1.8278301886792452,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004667298298810167,
      "loss": 5.4758,
      "mean_token_accuracy": 0.16594072431325912,
      "num_tokens": 41825763.0,
      "step": 20150
    },
    {
      "entropy": 5.724749898910522,
      "epoch": 1.8282837445573294,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004667128949437867,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1758809819817543,
      "num_tokens": 41834543.0,
      "step": 20155
    },
    {
      "entropy": 5.66655797958374,
      "epoch": 1.8287373004354137,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004666959560418852,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17227186113595963,
      "num_tokens": 41844940.0,
      "step": 20160
    },
    {
      "entropy": 5.611432600021362,
      "epoch": 1.829190856313498,
      "grad_norm": 7.40625,
      "learning_rate": 0.00046667901317566245,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17340007573366165,
      "num_tokens": 41855532.0,
      "step": 20165
    },
    {
      "entropy": 5.724791240692139,
      "epoch": 1.829644412191582,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004666620663454689,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1681271970272064,
      "num_tokens": 41865661.0,
      "step": 20170
    },
    {
      "entropy": 5.739521217346192,
      "epoch": 1.830097968069666,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046664511555165496,
      "loss": 5.4627,
      "mean_token_accuracy": 0.16760478913784027,
      "num_tokens": 41875662.0,
      "step": 20175
    },
    {
      "entropy": 5.72694149017334,
      "epoch": 1.8305515239477503,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046662816079457114,
      "loss": 5.4251,
      "mean_token_accuracy": 0.1794849932193756,
      "num_tokens": 41886211.0,
      "step": 20180
    },
    {
      "entropy": 5.798371934890747,
      "epoch": 1.8310050798258346,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004666112020745682,
      "loss": 5.5381,
      "mean_token_accuracy": 0.16386871933937072,
      "num_tokens": 41896737.0,
      "step": 20185
    },
    {
      "entropy": 5.774637031555176,
      "epoch": 1.8314586357039189,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004665942393919967,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1690554991364479,
      "num_tokens": 41907768.0,
      "step": 20190
    },
    {
      "entropy": 5.713705062866211,
      "epoch": 1.831912191582003,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046657727274720755,
      "loss": 5.3649,
      "mean_token_accuracy": 0.16781125664711,
      "num_tokens": 41917587.0,
      "step": 20195
    },
    {
      "entropy": 5.661926937103272,
      "epoch": 1.832365747460087,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004665603021405516,
      "loss": 5.3677,
      "mean_token_accuracy": 0.17560430914163588,
      "num_tokens": 41927842.0,
      "step": 20200
    },
    {
      "entropy": 5.7022803783416744,
      "epoch": 1.8328193033381712,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004665433275723798,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17469791918992997,
      "num_tokens": 41939899.0,
      "step": 20205
    },
    {
      "entropy": 5.6437889575958256,
      "epoch": 1.8332728592162555,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004665263490430431,
      "loss": 5.4184,
      "mean_token_accuracy": 0.16269447058439254,
      "num_tokens": 41950932.0,
      "step": 20210
    },
    {
      "entropy": 5.799565076828003,
      "epoch": 1.8337264150943398,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004665093665528928,
      "loss": 5.4932,
      "mean_token_accuracy": 0.17104425728321077,
      "num_tokens": 41963512.0,
      "step": 20215
    },
    {
      "entropy": 5.808188629150391,
      "epoch": 1.8341799709724238,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046649238010227983,
      "loss": 5.4805,
      "mean_token_accuracy": 0.1702400490641594,
      "num_tokens": 41974677.0,
      "step": 20220
    },
    {
      "entropy": 5.663489246368409,
      "epoch": 1.8346335268505078,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004664753896915558,
      "loss": 5.3753,
      "mean_token_accuracy": 0.17435685843229293,
      "num_tokens": 41984582.0,
      "step": 20225
    },
    {
      "entropy": 5.689115905761719,
      "epoch": 1.8350870827285921,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004664583953210719,
      "loss": 5.3665,
      "mean_token_accuracy": 0.1712415710091591,
      "num_tokens": 41994624.0,
      "step": 20230
    },
    {
      "entropy": 5.729493808746338,
      "epoch": 1.8355406386066764,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004664413969911795,
      "loss": 5.3841,
      "mean_token_accuracy": 0.1728738397359848,
      "num_tokens": 42004835.0,
      "step": 20235
    },
    {
      "entropy": 5.638566207885742,
      "epoch": 1.8359941944847606,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004664243947022303,
      "loss": 5.4054,
      "mean_token_accuracy": 0.17417764961719512,
      "num_tokens": 42014727.0,
      "step": 20240
    },
    {
      "entropy": 5.758063936233521,
      "epoch": 1.8364477503628447,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046640738845457586,
      "loss": 5.453,
      "mean_token_accuracy": 0.1711594581604004,
      "num_tokens": 42023628.0,
      "step": 20245
    },
    {
      "entropy": 5.791610383987427,
      "epoch": 1.8369013062409287,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004663903782485678,
      "loss": 5.4781,
      "mean_token_accuracy": 0.16456804275512696,
      "num_tokens": 42033587.0,
      "step": 20250
    },
    {
      "entropy": 5.75460205078125,
      "epoch": 1.837354862119013,
      "grad_norm": 5.65625,
      "learning_rate": 0.00046637336408455793,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16543664485216142,
      "num_tokens": 42043765.0,
      "step": 20255
    },
    {
      "entropy": 5.814415073394775,
      "epoch": 1.8378084179970973,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046635634596289816,
      "loss": 5.5333,
      "mean_token_accuracy": 0.16852532625198363,
      "num_tokens": 42054205.0,
      "step": 20260
    },
    {
      "entropy": 5.668474721908569,
      "epoch": 1.8382619738751815,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004663393238839404,
      "loss": 5.3332,
      "mean_token_accuracy": 0.16873594969511033,
      "num_tokens": 42064512.0,
      "step": 20265
    },
    {
      "entropy": 5.727361726760864,
      "epoch": 1.8387155297532656,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004663222978480366,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1683933198451996,
      "num_tokens": 42074954.0,
      "step": 20270
    },
    {
      "entropy": 5.7485860824584964,
      "epoch": 1.8391690856313496,
      "grad_norm": 1.78125,
      "learning_rate": 0.000466305267855539,
      "loss": 5.4543,
      "mean_token_accuracy": 0.16761189848184585,
      "num_tokens": 42084914.0,
      "step": 20275
    },
    {
      "entropy": 5.756715774536133,
      "epoch": 1.8396226415094339,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004662882339067996,
      "loss": 5.4742,
      "mean_token_accuracy": 0.17496540695428847,
      "num_tokens": 42095454.0,
      "step": 20280
    },
    {
      "entropy": 5.730036878585816,
      "epoch": 1.8400761973875182,
      "grad_norm": 4.25,
      "learning_rate": 0.0004662711960021709,
      "loss": 5.5248,
      "mean_token_accuracy": 0.1620542511343956,
      "num_tokens": 42106960.0,
      "step": 20285
    },
    {
      "entropy": 5.686562490463257,
      "epoch": 1.8405297532656024,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046625415414200513,
      "loss": 5.4253,
      "mean_token_accuracy": 0.16947820633649827,
      "num_tokens": 42117009.0,
      "step": 20290
    },
    {
      "entropy": 5.778832006454468,
      "epoch": 1.8409833091436865,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004662371083266546,
      "loss": 5.4178,
      "mean_token_accuracy": 0.17296259850263596,
      "num_tokens": 42126371.0,
      "step": 20295
    },
    {
      "entropy": 5.744422054290771,
      "epoch": 1.8414368650217707,
      "grad_norm": 2.34375,
      "learning_rate": 0.000466220058556472,
      "loss": 5.4497,
      "mean_token_accuracy": 0.1671949580311775,
      "num_tokens": 42135890.0,
      "step": 20300
    },
    {
      "entropy": 5.772117567062378,
      "epoch": 1.8418904208998548,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046620300483180986,
      "loss": 5.5581,
      "mean_token_accuracy": 0.16415327787399292,
      "num_tokens": 42145696.0,
      "step": 20305
    },
    {
      "entropy": 5.783477354049682,
      "epoch": 1.842343976777939,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004661859471530209,
      "loss": 5.4804,
      "mean_token_accuracy": 0.17040545344352723,
      "num_tokens": 42156236.0,
      "step": 20310
    },
    {
      "entropy": 5.775360918045044,
      "epoch": 1.8427975326560233,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004661688855204578,
      "loss": 5.4907,
      "mean_token_accuracy": 0.1700643017888069,
      "num_tokens": 42166179.0,
      "step": 20315
    },
    {
      "entropy": 5.680700445175171,
      "epoch": 1.8432510885341074,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046615181993447344,
      "loss": 5.366,
      "mean_token_accuracy": 0.16725947558879853,
      "num_tokens": 42176643.0,
      "step": 20320
    },
    {
      "entropy": 5.776334953308106,
      "epoch": 1.8437046444121916,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004661347503954207,
      "loss": 5.6258,
      "mean_token_accuracy": 0.15752689987421037,
      "num_tokens": 42188328.0,
      "step": 20325
    },
    {
      "entropy": 5.839367389678955,
      "epoch": 1.8441582002902757,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046611767690365265,
      "loss": 5.4707,
      "mean_token_accuracy": 0.1713878870010376,
      "num_tokens": 42198110.0,
      "step": 20330
    },
    {
      "entropy": 5.704930686950684,
      "epoch": 1.84461175616836,
      "grad_norm": 2.25,
      "learning_rate": 0.0004661005994595223,
      "loss": 5.4271,
      "mean_token_accuracy": 0.16396609842777252,
      "num_tokens": 42208438.0,
      "step": 20335
    },
    {
      "entropy": 5.727550649642945,
      "epoch": 1.8450653120464442,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004660835180633829,
      "loss": 5.4209,
      "mean_token_accuracy": 0.16924734115600587,
      "num_tokens": 42220052.0,
      "step": 20340
    },
    {
      "entropy": 5.696164941787719,
      "epoch": 1.8455188679245285,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004660664327155876,
      "loss": 5.4049,
      "mean_token_accuracy": 0.17349729388952256,
      "num_tokens": 42229497.0,
      "step": 20345
    },
    {
      "entropy": 5.690338563919068,
      "epoch": 1.8459724238026125,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004660493434164898,
      "loss": 5.4739,
      "mean_token_accuracy": 0.1604655936360359,
      "num_tokens": 42240725.0,
      "step": 20350
    },
    {
      "entropy": 5.773559331893921,
      "epoch": 1.8464259796806965,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004660322501664429,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1693146675825119,
      "num_tokens": 42251114.0,
      "step": 20355
    },
    {
      "entropy": 5.80815691947937,
      "epoch": 1.8468795355587808,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004660151529658004,
      "loss": 5.6406,
      "mean_token_accuracy": 0.16074103713035584,
      "num_tokens": 42262825.0,
      "step": 20360
    },
    {
      "entropy": 5.8122560501098635,
      "epoch": 1.847333091436865,
      "grad_norm": 6.65625,
      "learning_rate": 0.0004659980518149158,
      "loss": 5.4745,
      "mean_token_accuracy": 0.17571987509727477,
      "num_tokens": 42272176.0,
      "step": 20365
    },
    {
      "entropy": 5.805242824554443,
      "epoch": 1.8477866473149493,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046598094671414285,
      "loss": 5.4999,
      "mean_token_accuracy": 0.1704937696456909,
      "num_tokens": 42281934.0,
      "step": 20370
    },
    {
      "entropy": 5.730363082885742,
      "epoch": 1.8482402031930334,
      "grad_norm": 4.3125,
      "learning_rate": 0.00046596383766383533,
      "loss": 5.3747,
      "mean_token_accuracy": 0.17203826904296876,
      "num_tokens": 42291927.0,
      "step": 20375
    },
    {
      "entropy": 5.667574214935303,
      "epoch": 1.8486937590711174,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004659467246643469,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17182349413633347,
      "num_tokens": 42301407.0,
      "step": 20380
    },
    {
      "entropy": 5.634180164337158,
      "epoch": 1.8491473149492017,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046592960771603154,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1778141438961029,
      "num_tokens": 42311199.0,
      "step": 20385
    },
    {
      "entropy": 5.680311250686645,
      "epoch": 1.849600870827286,
      "grad_norm": 2.0,
      "learning_rate": 0.0004659124868192433,
      "loss": 5.3983,
      "mean_token_accuracy": 0.17388981431722642,
      "num_tokens": 42320725.0,
      "step": 20390
    },
    {
      "entropy": 5.802590799331665,
      "epoch": 1.8500544267053702,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004658953619743361,
      "loss": 5.4922,
      "mean_token_accuracy": 0.17125590294599533,
      "num_tokens": 42330857.0,
      "step": 20395
    },
    {
      "entropy": 5.699901533126831,
      "epoch": 1.8505079825834543,
      "grad_norm": 3.09375,
      "learning_rate": 0.00046587823318166427,
      "loss": 5.341,
      "mean_token_accuracy": 0.17103482633829117,
      "num_tokens": 42340241.0,
      "step": 20400
    },
    {
      "entropy": 5.74692063331604,
      "epoch": 1.8509615384615383,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046586110044158184,
      "loss": 5.5621,
      "mean_token_accuracy": 0.1607111796736717,
      "num_tokens": 42351756.0,
      "step": 20405
    },
    {
      "entropy": 5.71598105430603,
      "epoch": 1.8514150943396226,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046584396375444335,
      "loss": 5.3849,
      "mean_token_accuracy": 0.17296426445245744,
      "num_tokens": 42361649.0,
      "step": 20410
    },
    {
      "entropy": 5.717461442947387,
      "epoch": 1.8518686502177069,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046582682312060295,
      "loss": 5.4063,
      "mean_token_accuracy": 0.17577731162309645,
      "num_tokens": 42371273.0,
      "step": 20415
    },
    {
      "entropy": 5.701515769958496,
      "epoch": 1.8523222060957911,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004658096785404152,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1687558114528656,
      "num_tokens": 42382275.0,
      "step": 20420
    },
    {
      "entropy": 5.797797584533692,
      "epoch": 1.8527757619738752,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046579253001423473,
      "loss": 5.4567,
      "mean_token_accuracy": 0.16815458089113236,
      "num_tokens": 42392360.0,
      "step": 20425
    },
    {
      "entropy": 5.738983869552612,
      "epoch": 1.8532293178519592,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046577537754241604,
      "loss": 5.4719,
      "mean_token_accuracy": 0.1595483809709549,
      "num_tokens": 42403695.0,
      "step": 20430
    },
    {
      "entropy": 5.705842018127441,
      "epoch": 1.8536828737300435,
      "grad_norm": 3.671875,
      "learning_rate": 0.000465758221125314,
      "loss": 5.3835,
      "mean_token_accuracy": 0.17378616333007812,
      "num_tokens": 42414199.0,
      "step": 20435
    },
    {
      "entropy": 5.714840316772461,
      "epoch": 1.8541364296081277,
      "grad_norm": 3.5,
      "learning_rate": 0.00046574106076328333,
      "loss": 5.507,
      "mean_token_accuracy": 0.16989886313676833,
      "num_tokens": 42423655.0,
      "step": 20440
    },
    {
      "entropy": 5.7758995532989506,
      "epoch": 1.854589985486212,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004657238964566789,
      "loss": 5.4007,
      "mean_token_accuracy": 0.1737048089504242,
      "num_tokens": 42434235.0,
      "step": 20445
    },
    {
      "entropy": 5.704844999313354,
      "epoch": 1.855043541364296,
      "grad_norm": 2.625,
      "learning_rate": 0.0004657067282058557,
      "loss": 5.4348,
      "mean_token_accuracy": 0.16791796833276748,
      "num_tokens": 42444701.0,
      "step": 20450
    },
    {
      "entropy": 5.660160255432129,
      "epoch": 1.85549709724238,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004656895560111687,
      "loss": 5.4135,
      "mean_token_accuracy": 0.16817153841257096,
      "num_tokens": 42454499.0,
      "step": 20455
    },
    {
      "entropy": 5.775209951400757,
      "epoch": 1.8559506531204644,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046567237987297323,
      "loss": 5.3979,
      "mean_token_accuracy": 0.1604381814599037,
      "num_tokens": 42465391.0,
      "step": 20460
    },
    {
      "entropy": 5.819307136535644,
      "epoch": 1.8564042089985486,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004656551997916243,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16512527614831923,
      "num_tokens": 42476126.0,
      "step": 20465
    },
    {
      "entropy": 5.829487895965576,
      "epoch": 1.856857764876633,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046563801576747716,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16936197280883789,
      "num_tokens": 42486833.0,
      "step": 20470
    },
    {
      "entropy": 5.8151932716369625,
      "epoch": 1.857311320754717,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046562082780088733,
      "loss": 5.5044,
      "mean_token_accuracy": 0.1661322869360447,
      "num_tokens": 42496986.0,
      "step": 20475
    },
    {
      "entropy": 5.701701498031616,
      "epoch": 1.8577648766328012,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004656036358922103,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1693151354789734,
      "num_tokens": 42507352.0,
      "step": 20480
    },
    {
      "entropy": 5.723337459564209,
      "epoch": 1.8582184325108853,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004655864400418014,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1716725930571556,
      "num_tokens": 42517655.0,
      "step": 20485
    },
    {
      "entropy": 5.732774925231934,
      "epoch": 1.8586719883889695,
      "grad_norm": 1.75,
      "learning_rate": 0.0004655692402500163,
      "loss": 5.3649,
      "mean_token_accuracy": 0.18143543004989623,
      "num_tokens": 42527502.0,
      "step": 20490
    },
    {
      "entropy": 5.72069034576416,
      "epoch": 1.8591255442670538,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046555203651721085,
      "loss": 5.4106,
      "mean_token_accuracy": 0.17047982662916183,
      "num_tokens": 42538893.0,
      "step": 20495
    },
    {
      "entropy": 5.575335454940796,
      "epoch": 1.8595791001451378,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004655348288437407,
      "loss": 5.298,
      "mean_token_accuracy": 0.1818207561969757,
      "num_tokens": 42550372.0,
      "step": 20500
    },
    {
      "entropy": 5.760228252410888,
      "epoch": 1.860032656023222,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004655176172299618,
      "loss": 5.532,
      "mean_token_accuracy": 0.16528619229793548,
      "num_tokens": 42560814.0,
      "step": 20505
    },
    {
      "entropy": 5.784014272689819,
      "epoch": 1.8604862119013061,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046550040167622986,
      "loss": 5.5117,
      "mean_token_accuracy": 0.16645128577947615,
      "num_tokens": 42570993.0,
      "step": 20510
    },
    {
      "entropy": 5.754129409790039,
      "epoch": 1.8609397677793904,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046548318218290124,
      "loss": 5.3685,
      "mean_token_accuracy": 0.17316617518663407,
      "num_tokens": 42581357.0,
      "step": 20515
    },
    {
      "entropy": 5.736874151229858,
      "epoch": 1.8613933236574747,
      "grad_norm": 5.625,
      "learning_rate": 0.0004654659587503318,
      "loss": 5.473,
      "mean_token_accuracy": 0.16084801703691481,
      "num_tokens": 42591653.0,
      "step": 20520
    },
    {
      "entropy": 5.750981998443604,
      "epoch": 1.861846879535559,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004654487313788778,
      "loss": 5.5218,
      "mean_token_accuracy": 0.16863651275634767,
      "num_tokens": 42602232.0,
      "step": 20525
    },
    {
      "entropy": 5.738296031951904,
      "epoch": 1.862300435413643,
      "grad_norm": 3.390625,
      "learning_rate": 0.00046543150006889546,
      "loss": 5.4648,
      "mean_token_accuracy": 0.17337899655103683,
      "num_tokens": 42613283.0,
      "step": 20530
    },
    {
      "entropy": 5.754702043533325,
      "epoch": 1.862753991291727,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004654142648207412,
      "loss": 5.3808,
      "mean_token_accuracy": 0.17422992289066314,
      "num_tokens": 42623594.0,
      "step": 20535
    },
    {
      "entropy": 5.760178279876709,
      "epoch": 1.8632075471698113,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004653970256347715,
      "loss": 5.4272,
      "mean_token_accuracy": 0.16677011400461197,
      "num_tokens": 42633587.0,
      "step": 20540
    },
    {
      "entropy": 5.710048961639404,
      "epoch": 1.8636611030478956,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046537978251134265,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1673750787973404,
      "num_tokens": 42643426.0,
      "step": 20545
    },
    {
      "entropy": 5.645929145812988,
      "epoch": 1.8641146589259798,
      "grad_norm": 4.0,
      "learning_rate": 0.0004653625354508115,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16745571047067642,
      "num_tokens": 42653521.0,
      "step": 20550
    },
    {
      "entropy": 5.824835014343262,
      "epoch": 1.8645682148040639,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004653452844535346,
      "loss": 5.4728,
      "mean_token_accuracy": 0.16896266043186187,
      "num_tokens": 42664818.0,
      "step": 20555
    },
    {
      "entropy": 5.811244201660156,
      "epoch": 1.865021770682148,
      "grad_norm": 3.203125,
      "learning_rate": 0.00046532802951986863,
      "loss": 5.48,
      "mean_token_accuracy": 0.17234469801187516,
      "num_tokens": 42674738.0,
      "step": 20560
    },
    {
      "entropy": 5.684953212738037,
      "epoch": 1.8654753265602322,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004653107706501705,
      "loss": 5.4005,
      "mean_token_accuracy": 0.1803669661283493,
      "num_tokens": 42685003.0,
      "step": 20565
    },
    {
      "entropy": 5.657580757141114,
      "epoch": 1.8659288824383164,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004652935078447973,
      "loss": 5.4606,
      "mean_token_accuracy": 0.17217807471752167,
      "num_tokens": 42695428.0,
      "step": 20570
    },
    {
      "entropy": 5.80774621963501,
      "epoch": 1.8663824383164007,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046527624110410574,
      "loss": 5.3969,
      "mean_token_accuracy": 0.17665866166353225,
      "num_tokens": 42704901.0,
      "step": 20575
    },
    {
      "entropy": 5.785603952407837,
      "epoch": 1.8668359941944848,
      "grad_norm": 2.8125,
      "learning_rate": 0.000465258970428453,
      "loss": 5.4929,
      "mean_token_accuracy": 0.16307479590177537,
      "num_tokens": 42714379.0,
      "step": 20580
    },
    {
      "entropy": 5.797417449951172,
      "epoch": 1.8672895500725688,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046524169581819633,
      "loss": 5.5964,
      "mean_token_accuracy": 0.16371024698019027,
      "num_tokens": 42724813.0,
      "step": 20585
    },
    {
      "entropy": 5.734457731246948,
      "epoch": 1.867743105950653,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004652244172736929,
      "loss": 5.4344,
      "mean_token_accuracy": 0.16969114392995835,
      "num_tokens": 42736590.0,
      "step": 20590
    },
    {
      "entropy": 5.760493326187134,
      "epoch": 1.8681966618287373,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004652071347953,
      "loss": 5.5119,
      "mean_token_accuracy": 0.16458907425403596,
      "num_tokens": 42746718.0,
      "step": 20595
    },
    {
      "entropy": 5.736437702178955,
      "epoch": 1.8686502177068216,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046518984838337513,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16649200171232223,
      "num_tokens": 42756484.0,
      "step": 20600
    },
    {
      "entropy": 5.774122858047486,
      "epoch": 1.8691037735849056,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004651725580382757,
      "loss": 5.5202,
      "mean_token_accuracy": 0.16852520555257797,
      "num_tokens": 42768512.0,
      "step": 20605
    },
    {
      "entropy": 5.6991160869598385,
      "epoch": 1.8695573294629897,
      "grad_norm": 3.75,
      "learning_rate": 0.00046515526376035934,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17365105003118514,
      "num_tokens": 42779812.0,
      "step": 20610
    },
    {
      "entropy": 5.632942771911621,
      "epoch": 1.870010885341074,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004651379655499837,
      "loss": 5.3773,
      "mean_token_accuracy": 0.16976198107004165,
      "num_tokens": 42791163.0,
      "step": 20615
    },
    {
      "entropy": 5.668783473968506,
      "epoch": 1.8704644412191582,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046512066340750636,
      "loss": 5.4427,
      "mean_token_accuracy": 0.16669269651174545,
      "num_tokens": 42801810.0,
      "step": 20620
    },
    {
      "entropy": 5.807477331161499,
      "epoch": 1.8709179970972425,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004651033573332853,
      "loss": 5.4691,
      "mean_token_accuracy": 0.17028764635324478,
      "num_tokens": 42813358.0,
      "step": 20625
    },
    {
      "entropy": 5.6463428974151615,
      "epoch": 1.8713715529753265,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004650860473276784,
      "loss": 5.2761,
      "mean_token_accuracy": 0.18685212582349778,
      "num_tokens": 42823519.0,
      "step": 20630
    },
    {
      "entropy": 5.687486696243286,
      "epoch": 1.8718251088534106,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004650687333910436,
      "loss": 5.4389,
      "mean_token_accuracy": 0.17469302713871002,
      "num_tokens": 42833257.0,
      "step": 20635
    },
    {
      "entropy": 5.760562944412231,
      "epoch": 1.8722786647314948,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046505141552373897,
      "loss": 5.511,
      "mean_token_accuracy": 0.1661946251988411,
      "num_tokens": 42843536.0,
      "step": 20640
    },
    {
      "entropy": 5.702634382247925,
      "epoch": 1.8727322206095791,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004650340937261226,
      "loss": 5.4101,
      "mean_token_accuracy": 0.17409064769744872,
      "num_tokens": 42853911.0,
      "step": 20645
    },
    {
      "entropy": 5.783473968505859,
      "epoch": 1.8731857764876634,
      "grad_norm": 2.5,
      "learning_rate": 0.0004650167679985528,
      "loss": 5.5311,
      "mean_token_accuracy": 0.16294777542352676,
      "num_tokens": 42864173.0,
      "step": 20650
    },
    {
      "entropy": 5.751931524276733,
      "epoch": 1.8736393323657474,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004649994383413877,
      "loss": 5.3756,
      "mean_token_accuracy": 0.18061547577381135,
      "num_tokens": 42873470.0,
      "step": 20655
    },
    {
      "entropy": 5.685538196563721,
      "epoch": 1.8740928882438317,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046498210475498595,
      "loss": 5.371,
      "mean_token_accuracy": 0.17386945635080336,
      "num_tokens": 42882496.0,
      "step": 20660
    },
    {
      "entropy": 5.679009819030762,
      "epoch": 1.8745464441219157,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046496476723970574,
      "loss": 5.3802,
      "mean_token_accuracy": 0.1734359323978424,
      "num_tokens": 42893163.0,
      "step": 20665
    },
    {
      "entropy": 5.710430765151978,
      "epoch": 1.875,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004649474257959058,
      "loss": 5.4334,
      "mean_token_accuracy": 0.16961780190467834,
      "num_tokens": 42903406.0,
      "step": 20670
    },
    {
      "entropy": 5.7534136295318605,
      "epoch": 1.8754535558780843,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004649300804239447,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1705908328294754,
      "num_tokens": 42913702.0,
      "step": 20675
    },
    {
      "entropy": 5.7026591300964355,
      "epoch": 1.8759071117561683,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046491273112418105,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17132111638784409,
      "num_tokens": 42923877.0,
      "step": 20680
    },
    {
      "entropy": 5.675516271591187,
      "epoch": 1.8763606676342526,
      "grad_norm": 3.0625,
      "learning_rate": 0.00046489537789697386,
      "loss": 5.3932,
      "mean_token_accuracy": 0.17618172764778137,
      "num_tokens": 42934393.0,
      "step": 20685
    },
    {
      "entropy": 5.770949268341065,
      "epoch": 1.8768142235123366,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046487802074268177,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16890577375888824,
      "num_tokens": 42945803.0,
      "step": 20690
    },
    {
      "entropy": 5.71511869430542,
      "epoch": 1.8772677793904209,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046486065966166377,
      "loss": 5.403,
      "mean_token_accuracy": 0.17555399984121323,
      "num_tokens": 42955829.0,
      "step": 20695
    },
    {
      "entropy": 5.725791501998901,
      "epoch": 1.8777213352685052,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004648432946542791,
      "loss": 5.465,
      "mean_token_accuracy": 0.16536943465471268,
      "num_tokens": 42967080.0,
      "step": 20700
    },
    {
      "entropy": 5.700277185440063,
      "epoch": 1.8781748911465894,
      "grad_norm": 4.375,
      "learning_rate": 0.00046482592572088654,
      "loss": 5.4252,
      "mean_token_accuracy": 0.16470708400011064,
      "num_tokens": 42978082.0,
      "step": 20705
    },
    {
      "entropy": 5.737454080581665,
      "epoch": 1.8786284470246735,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046480855286184565,
      "loss": 5.4939,
      "mean_token_accuracy": 0.16664159446954727,
      "num_tokens": 42988982.0,
      "step": 20710
    },
    {
      "entropy": 5.724548768997193,
      "epoch": 1.8790820029027575,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046479117607751535,
      "loss": 5.4417,
      "mean_token_accuracy": 0.16695476323366165,
      "num_tokens": 42998982.0,
      "step": 20715
    },
    {
      "entropy": 5.7334184646606445,
      "epoch": 1.8795355587808418,
      "grad_norm": 2.125,
      "learning_rate": 0.00046477379536825534,
      "loss": 5.3723,
      "mean_token_accuracy": 0.1754404440522194,
      "num_tokens": 43009574.0,
      "step": 20720
    },
    {
      "entropy": 5.726659965515137,
      "epoch": 1.879989114658926,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004647564107344247,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16645280122756959,
      "num_tokens": 43019335.0,
      "step": 20725
    },
    {
      "entropy": 5.657132768630982,
      "epoch": 1.8804426705370103,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046473902217638327,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1725488632917404,
      "num_tokens": 43029264.0,
      "step": 20730
    },
    {
      "entropy": 5.706268548965454,
      "epoch": 1.8808962264150944,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046472162969449037,
      "loss": 5.4357,
      "mean_token_accuracy": 0.17212045639753343,
      "num_tokens": 43040519.0,
      "step": 20735
    },
    {
      "entropy": 5.755479383468628,
      "epoch": 1.8813497822931784,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046470423328910594,
      "loss": 5.4428,
      "mean_token_accuracy": 0.1725601240992546,
      "num_tokens": 43051146.0,
      "step": 20740
    },
    {
      "entropy": 5.778139114379883,
      "epoch": 1.8818033381712627,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046468683296058957,
      "loss": 5.5307,
      "mean_token_accuracy": 0.1672705590724945,
      "num_tokens": 43061339.0,
      "step": 20745
    },
    {
      "entropy": 5.721444225311279,
      "epoch": 1.882256894049347,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046466942870930126,
      "loss": 5.4442,
      "mean_token_accuracy": 0.17231858372688294,
      "num_tokens": 43072624.0,
      "step": 20750
    },
    {
      "entropy": 5.698532676696777,
      "epoch": 1.8827104499274312,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004646520205356007,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17262689769268036,
      "num_tokens": 43082120.0,
      "step": 20755
    },
    {
      "entropy": 5.796448373794556,
      "epoch": 1.8831640058055152,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046463460843984805,
      "loss": 5.5226,
      "mean_token_accuracy": 0.16351934522390366,
      "num_tokens": 43093202.0,
      "step": 20760
    },
    {
      "entropy": 5.892607307434082,
      "epoch": 1.8836175616835993,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004646171924224034,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1672188714146614,
      "num_tokens": 43104801.0,
      "step": 20765
    },
    {
      "entropy": 5.74223313331604,
      "epoch": 1.8840711175616836,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046459977248362687,
      "loss": 5.3976,
      "mean_token_accuracy": 0.17323997914791106,
      "num_tokens": 43114855.0,
      "step": 20770
    },
    {
      "entropy": 5.675221347808838,
      "epoch": 1.8845246734397678,
      "grad_norm": 2.0,
      "learning_rate": 0.0004645823486238788,
      "loss": 5.4032,
      "mean_token_accuracy": 0.16519128382205964,
      "num_tokens": 43124759.0,
      "step": 20775
    },
    {
      "entropy": 5.737265968322754,
      "epoch": 1.884978229317852,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004645649208435193,
      "loss": 5.3938,
      "mean_token_accuracy": 0.1756272107362747,
      "num_tokens": 43135526.0,
      "step": 20780
    },
    {
      "entropy": 5.73284878730774,
      "epoch": 1.8854317851959361,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046454748914290906,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1667809620499611,
      "num_tokens": 43145605.0,
      "step": 20785
    },
    {
      "entropy": 5.767112731933594,
      "epoch": 1.8858853410740202,
      "grad_norm": 2.375,
      "learning_rate": 0.00046453005352240835,
      "loss": 5.5645,
      "mean_token_accuracy": 0.16321603953838348,
      "num_tokens": 43156578.0,
      "step": 20790
    },
    {
      "entropy": 5.840833520889282,
      "epoch": 1.8863388969521044,
      "grad_norm": 2.453125,
      "learning_rate": 0.000464512613982378,
      "loss": 5.4314,
      "mean_token_accuracy": 0.17289584279060363,
      "num_tokens": 43166870.0,
      "step": 20795
    },
    {
      "entropy": 5.67694501876831,
      "epoch": 1.8867924528301887,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004644951705231783,
      "loss": 5.3948,
      "mean_token_accuracy": 0.17385641038417815,
      "num_tokens": 43176406.0,
      "step": 20800
    },
    {
      "entropy": 5.7598662853240965,
      "epoch": 1.887246008708273,
      "grad_norm": 2.375,
      "learning_rate": 0.00046447772314517024,
      "loss": 5.5964,
      "mean_token_accuracy": 0.16601465344429017,
      "num_tokens": 43187502.0,
      "step": 20805
    },
    {
      "entropy": 5.7123894691467285,
      "epoch": 1.887699564586357,
      "grad_norm": 2.765625,
      "learning_rate": 0.00046446027184871464,
      "loss": 5.348,
      "mean_token_accuracy": 0.17434955388307571,
      "num_tokens": 43197928.0,
      "step": 20810
    },
    {
      "entropy": 5.740464878082276,
      "epoch": 1.888153120464441,
      "grad_norm": 2.859375,
      "learning_rate": 0.00046444281663417236,
      "loss": 5.4546,
      "mean_token_accuracy": 0.17700834423303605,
      "num_tokens": 43209083.0,
      "step": 20815
    },
    {
      "entropy": 5.693803596496582,
      "epoch": 1.8886066763425253,
      "grad_norm": 3.5,
      "learning_rate": 0.0004644253575019043,
      "loss": 5.4227,
      "mean_token_accuracy": 0.16881635487079621,
      "num_tokens": 43219089.0,
      "step": 20820
    },
    {
      "entropy": 5.7675432682037355,
      "epoch": 1.8890602322206096,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004644078944522716,
      "loss": 5.4368,
      "mean_token_accuracy": 0.17267759144306183,
      "num_tokens": 43229440.0,
      "step": 20825
    },
    {
      "entropy": 5.7658665657043455,
      "epoch": 1.8895137880986939,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004643904274856353,
      "loss": 5.427,
      "mean_token_accuracy": 0.1693614423274994,
      "num_tokens": 43239995.0,
      "step": 20830
    },
    {
      "entropy": 5.707099485397339,
      "epoch": 1.889967343976778,
      "grad_norm": 3.71875,
      "learning_rate": 0.00046437295660235674,
      "loss": 5.4052,
      "mean_token_accuracy": 0.17868402004241943,
      "num_tokens": 43249722.0,
      "step": 20835
    },
    {
      "entropy": 5.670976066589356,
      "epoch": 1.8904208998548622,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004643554818027972,
      "loss": 5.4541,
      "mean_token_accuracy": 0.17103859037160873,
      "num_tokens": 43260029.0,
      "step": 20840
    },
    {
      "entropy": 5.7347540855407715,
      "epoch": 1.8908744557329462,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004643380030873181,
      "loss": 5.424,
      "mean_token_accuracy": 0.17188406139612197,
      "num_tokens": 43270571.0,
      "step": 20845
    },
    {
      "entropy": 5.710753011703491,
      "epoch": 1.8913280116110305,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004643205204562808,
      "loss": 5.4009,
      "mean_token_accuracy": 0.1736111342906952,
      "num_tokens": 43281310.0,
      "step": 20850
    },
    {
      "entropy": 5.754141330718994,
      "epoch": 1.8917815674891147,
      "grad_norm": 2.375,
      "learning_rate": 0.00046430303391004687,
      "loss": 5.444,
      "mean_token_accuracy": 0.17490168064832687,
      "num_tokens": 43292904.0,
      "step": 20855
    },
    {
      "entropy": 5.766540050506592,
      "epoch": 1.8922351233671988,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004642855434489779,
      "loss": 5.4125,
      "mean_token_accuracy": 0.17096040695905684,
      "num_tokens": 43303745.0,
      "step": 20860
    },
    {
      "entropy": 5.702844953536987,
      "epoch": 1.892688679245283,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046426804907343574,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16999629139900208,
      "num_tokens": 43314652.0,
      "step": 20865
    },
    {
      "entropy": 5.678434228897094,
      "epoch": 1.893142235123367,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046425055078378206,
      "loss": 5.3181,
      "mean_token_accuracy": 0.18321368992328643,
      "num_tokens": 43324242.0,
      "step": 20870
    },
    {
      "entropy": 5.652900457382202,
      "epoch": 1.8935957910014514,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046423304858037885,
      "loss": 5.3974,
      "mean_token_accuracy": 0.18353109508752824,
      "num_tokens": 43334054.0,
      "step": 20875
    },
    {
      "entropy": 5.657629346847534,
      "epoch": 1.8940493468795356,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004642155424635878,
      "loss": 5.3195,
      "mean_token_accuracy": 0.18116371035575868,
      "num_tokens": 43343389.0,
      "step": 20880
    },
    {
      "entropy": 5.668998289108276,
      "epoch": 1.89450290275762,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046419803243377125,
      "loss": 5.4026,
      "mean_token_accuracy": 0.17645146548748017,
      "num_tokens": 43353892.0,
      "step": 20885
    },
    {
      "entropy": 5.6746519088745115,
      "epoch": 1.894956458635704,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046418051849129115,
      "loss": 5.4102,
      "mean_token_accuracy": 0.1767694279551506,
      "num_tokens": 43363912.0,
      "step": 20890
    },
    {
      "entropy": 5.728085470199585,
      "epoch": 1.895410014513788,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004641630006365097,
      "loss": 5.4833,
      "mean_token_accuracy": 0.16770465821027755,
      "num_tokens": 43374719.0,
      "step": 20895
    },
    {
      "entropy": 5.746254253387451,
      "epoch": 1.8958635703918723,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046414547886978926,
      "loss": 5.4719,
      "mean_token_accuracy": 0.17165803015232087,
      "num_tokens": 43385549.0,
      "step": 20900
    },
    {
      "entropy": 5.737391567230224,
      "epoch": 1.8963171262699565,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046412795319149204,
      "loss": 5.4806,
      "mean_token_accuracy": 0.17429056018590927,
      "num_tokens": 43395503.0,
      "step": 20905
    },
    {
      "entropy": 5.760853672027588,
      "epoch": 1.8967706821480408,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046411042360198053,
      "loss": 5.4772,
      "mean_token_accuracy": 0.1598276674747467,
      "num_tokens": 43406913.0,
      "step": 20910
    },
    {
      "entropy": 5.759226369857788,
      "epoch": 1.8972242380261248,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004640928901016173,
      "loss": 5.3643,
      "mean_token_accuracy": 0.17044385969638826,
      "num_tokens": 43416459.0,
      "step": 20915
    },
    {
      "entropy": 5.6472580909729,
      "epoch": 1.8976777939042089,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046407535269076493,
      "loss": 5.4388,
      "mean_token_accuracy": 0.16799082159996032,
      "num_tokens": 43426280.0,
      "step": 20920
    },
    {
      "entropy": 5.816014337539673,
      "epoch": 1.8981313497822931,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046405781136978606,
      "loss": 5.4917,
      "mean_token_accuracy": 0.17143782377243041,
      "num_tokens": 43436676.0,
      "step": 20925
    },
    {
      "entropy": 5.874925661087036,
      "epoch": 1.8985849056603774,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046404026613904357,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16424358040094375,
      "num_tokens": 43447019.0,
      "step": 20930
    },
    {
      "entropy": 5.6975923538208,
      "epoch": 1.8990384615384617,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004640227169989001,
      "loss": 5.42,
      "mean_token_accuracy": 0.17019583582878112,
      "num_tokens": 43456609.0,
      "step": 20935
    },
    {
      "entropy": 5.68717679977417,
      "epoch": 1.8994920174165457,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004640051639497187,
      "loss": 5.4748,
      "mean_token_accuracy": 0.16933676451444626,
      "num_tokens": 43467616.0,
      "step": 20940
    },
    {
      "entropy": 5.730819845199585,
      "epoch": 1.8999455732946298,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004639876069918624,
      "loss": 5.46,
      "mean_token_accuracy": 0.16744293123483658,
      "num_tokens": 43478209.0,
      "step": 20945
    },
    {
      "entropy": 5.749806594848633,
      "epoch": 1.900399129172714,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046397004612569416,
      "loss": 5.33,
      "mean_token_accuracy": 0.1733709454536438,
      "num_tokens": 43487278.0,
      "step": 20950
    },
    {
      "entropy": 5.648739385604858,
      "epoch": 1.9008526850507983,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046395248135157715,
      "loss": 5.3786,
      "mean_token_accuracy": 0.17498134076595306,
      "num_tokens": 43497457.0,
      "step": 20955
    },
    {
      "entropy": 5.682187557220459,
      "epoch": 1.9013062409288826,
      "grad_norm": 2.375,
      "learning_rate": 0.00046393491266987484,
      "loss": 5.4734,
      "mean_token_accuracy": 0.16897646337747574,
      "num_tokens": 43507761.0,
      "step": 20960
    },
    {
      "entropy": 5.646828937530517,
      "epoch": 1.9017597968069666,
      "grad_norm": 2.5,
      "learning_rate": 0.0004639173400809502,
      "loss": 5.3863,
      "mean_token_accuracy": 0.16744173765182496,
      "num_tokens": 43517198.0,
      "step": 20965
    },
    {
      "entropy": 5.745282554626465,
      "epoch": 1.9022133526850507,
      "grad_norm": 3.75,
      "learning_rate": 0.0004638997635851669,
      "loss": 5.4845,
      "mean_token_accuracy": 0.16608664244413376,
      "num_tokens": 43527429.0,
      "step": 20970
    },
    {
      "entropy": 5.863715505599975,
      "epoch": 1.902666908563135,
      "grad_norm": 5.96875,
      "learning_rate": 0.00046388218318288844,
      "loss": 5.493,
      "mean_token_accuracy": 0.1770082727074623,
      "num_tokens": 43537867.0,
      "step": 20975
    },
    {
      "entropy": 5.747127437591553,
      "epoch": 1.9031204644412192,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004638645988744782,
      "loss": 5.433,
      "mean_token_accuracy": 0.1701931193470955,
      "num_tokens": 43547284.0,
      "step": 20980
    },
    {
      "entropy": 5.738501405715942,
      "epoch": 1.9035740203193035,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004638470106603,
      "loss": 5.4788,
      "mean_token_accuracy": 0.1689433440566063,
      "num_tokens": 43557693.0,
      "step": 20985
    },
    {
      "entropy": 5.719828748703003,
      "epoch": 1.9040275761973875,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004638294185407174,
      "loss": 5.4151,
      "mean_token_accuracy": 0.178651162981987,
      "num_tokens": 43566909.0,
      "step": 20990
    },
    {
      "entropy": 5.718988561630249,
      "epoch": 1.9044811320754715,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004638118225160944,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17440591305494307,
      "num_tokens": 43576963.0,
      "step": 20995
    },
    {
      "entropy": 5.695768642425537,
      "epoch": 1.9049346879535558,
      "grad_norm": 3.046875,
      "learning_rate": 0.00046379422258679477,
      "loss": 5.4107,
      "mean_token_accuracy": 0.16871645152568818,
      "num_tokens": 43586649.0,
      "step": 21000
    },
    {
      "epoch": 1.9049346879535558,
      "eval_entropy": 5.491105280095096,
      "eval_loss": 5.49411153793335,
      "eval_mean_token_accuracy": 0.1752989698108707,
      "eval_num_tokens": 43586649.0,
      "eval_runtime": 26.576,
      "eval_samples_per_second": 1330.525,
      "eval_steps_per_second": 166.316,
      "step": 21000
    },
    {
      "entropy": 5.680195140838623,
      "epoch": 1.90538824383164,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004637766187531825,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1724276974797249,
      "num_tokens": 43596506.0,
      "step": 21005
    },
    {
      "entropy": 5.743009567260742,
      "epoch": 1.9058417997097243,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046375901101562163,
      "loss": 5.4512,
      "mean_token_accuracy": 0.17103558778762817,
      "num_tokens": 43607188.0,
      "step": 21010
    },
    {
      "entropy": 5.724486017227173,
      "epoch": 1.9062953555878084,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004637413993744763,
      "loss": 5.3902,
      "mean_token_accuracy": 0.1791769951581955,
      "num_tokens": 43617465.0,
      "step": 21015
    },
    {
      "entropy": 5.692719030380249,
      "epoch": 1.9067489114658926,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046372378383011077,
      "loss": 5.3838,
      "mean_token_accuracy": 0.17591278105974198,
      "num_tokens": 43627692.0,
      "step": 21020
    },
    {
      "entropy": 5.715708017349243,
      "epoch": 1.9072024673439767,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046370616438288933,
      "loss": 5.4868,
      "mean_token_accuracy": 0.1663670316338539,
      "num_tokens": 43637350.0,
      "step": 21025
    },
    {
      "entropy": 5.717444515228271,
      "epoch": 1.907656023222061,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046368854103317624,
      "loss": 5.4423,
      "mean_token_accuracy": 0.17051386833190918,
      "num_tokens": 43647092.0,
      "step": 21030
    },
    {
      "entropy": 5.729173946380615,
      "epoch": 1.9081095791001452,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046367091378133615,
      "loss": 5.424,
      "mean_token_accuracy": 0.17650021910667418,
      "num_tokens": 43657237.0,
      "step": 21035
    },
    {
      "entropy": 5.723619747161865,
      "epoch": 1.9085631349782293,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004636532826277333,
      "loss": 5.44,
      "mean_token_accuracy": 0.17128027379512786,
      "num_tokens": 43667966.0,
      "step": 21040
    },
    {
      "entropy": 5.668895244598389,
      "epoch": 1.9090166908563135,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046363564757273264,
      "loss": 5.4071,
      "mean_token_accuracy": 0.17222215086221696,
      "num_tokens": 43678059.0,
      "step": 21045
    },
    {
      "entropy": 5.717968606948853,
      "epoch": 1.9094702467343976,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046361800861669867,
      "loss": 5.4323,
      "mean_token_accuracy": 0.17552920281887055,
      "num_tokens": 43687556.0,
      "step": 21050
    },
    {
      "entropy": 5.777753591537476,
      "epoch": 1.9099238026124818,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046360036575999623,
      "loss": 5.4403,
      "mean_token_accuracy": 0.17006113529205322,
      "num_tokens": 43697791.0,
      "step": 21055
    },
    {
      "entropy": 5.663078260421753,
      "epoch": 1.9103773584905661,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004635827190029902,
      "loss": 5.3437,
      "mean_token_accuracy": 0.16892659068107604,
      "num_tokens": 43707390.0,
      "step": 21060
    },
    {
      "entropy": 5.789041948318482,
      "epoch": 1.9108309143686504,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004635650683460454,
      "loss": 5.4876,
      "mean_token_accuracy": 0.1694726213812828,
      "num_tokens": 43716842.0,
      "step": 21065
    },
    {
      "entropy": 5.70715560913086,
      "epoch": 1.9112844702467344,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000463547413789527,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17070740461349487,
      "num_tokens": 43726967.0,
      "step": 21070
    },
    {
      "entropy": 5.71885051727295,
      "epoch": 1.9117380261248185,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046352975533379997,
      "loss": 5.4284,
      "mean_token_accuracy": 0.17538302391767502,
      "num_tokens": 43737425.0,
      "step": 21075
    },
    {
      "entropy": 5.627499341964722,
      "epoch": 1.9121915820029027,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046351209297922964,
      "loss": 5.3793,
      "mean_token_accuracy": 0.17565181255340576,
      "num_tokens": 43747395.0,
      "step": 21080
    },
    {
      "entropy": 5.734213352203369,
      "epoch": 1.912645137880987,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046349442672618114,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16874746084213257,
      "num_tokens": 43756348.0,
      "step": 21085
    },
    {
      "entropy": 5.724384784698486,
      "epoch": 1.9130986937590713,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046347675657501986,
      "loss": 5.4126,
      "mean_token_accuracy": 0.17085857540369034,
      "num_tokens": 43767629.0,
      "step": 21090
    },
    {
      "entropy": 5.691259384155273,
      "epoch": 1.9135522496371553,
      "grad_norm": 1.625,
      "learning_rate": 0.0004634590825261113,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17996491938829423,
      "num_tokens": 43777398.0,
      "step": 21095
    },
    {
      "entropy": 5.756093883514405,
      "epoch": 1.9140058055152394,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046344140457982075,
      "loss": 5.4653,
      "mean_token_accuracy": 0.17022579163312912,
      "num_tokens": 43787605.0,
      "step": 21100
    },
    {
      "entropy": 5.6845355987548825,
      "epoch": 1.9144593613933236,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000463423722736514,
      "loss": 5.4265,
      "mean_token_accuracy": 0.1702544867992401,
      "num_tokens": 43796989.0,
      "step": 21105
    },
    {
      "entropy": 5.70713095664978,
      "epoch": 1.9149129172714079,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004634060369965567,
      "loss": 5.4338,
      "mean_token_accuracy": 0.1697176530957222,
      "num_tokens": 43806653.0,
      "step": 21110
    },
    {
      "entropy": 5.684105348587036,
      "epoch": 1.9153664731494922,
      "grad_norm": 2.125,
      "learning_rate": 0.00046338834736031446,
      "loss": 5.3446,
      "mean_token_accuracy": 0.17211282402276992,
      "num_tokens": 43817152.0,
      "step": 21115
    },
    {
      "entropy": 5.756568813323975,
      "epoch": 1.9158200290275762,
      "grad_norm": 2.25,
      "learning_rate": 0.00046337065382815324,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17162238657474518,
      "num_tokens": 43827341.0,
      "step": 21120
    },
    {
      "entropy": 5.638355159759522,
      "epoch": 1.9162735849056602,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046335295640043887,
      "loss": 5.3774,
      "mean_token_accuracy": 0.18042928725481033,
      "num_tokens": 43837195.0,
      "step": 21125
    },
    {
      "entropy": 5.652200412750244,
      "epoch": 1.9167271407837445,
      "grad_norm": 2.5,
      "learning_rate": 0.00046333525507753743,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17460718750953674,
      "num_tokens": 43847473.0,
      "step": 21130
    },
    {
      "entropy": 5.645385503768921,
      "epoch": 1.9171806966618288,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004633175498598148,
      "loss": 5.3851,
      "mean_token_accuracy": 0.17354338467121125,
      "num_tokens": 43857590.0,
      "step": 21135
    },
    {
      "entropy": 5.738726282119751,
      "epoch": 1.917634252539913,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046329984074763733,
      "loss": 5.518,
      "mean_token_accuracy": 0.16866867244243622,
      "num_tokens": 43867085.0,
      "step": 21140
    },
    {
      "entropy": 5.722367811203003,
      "epoch": 1.918087808417997,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004632821277413711,
      "loss": 5.3713,
      "mean_token_accuracy": 0.1715300887823105,
      "num_tokens": 43877153.0,
      "step": 21145
    },
    {
      "entropy": 5.78055772781372,
      "epoch": 1.9185413642960811,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004632644108413825,
      "loss": 5.4971,
      "mean_token_accuracy": 0.15384760051965712,
      "num_tokens": 43887977.0,
      "step": 21150
    },
    {
      "entropy": 5.69014949798584,
      "epoch": 1.9189949201741654,
      "grad_norm": 3.3125,
      "learning_rate": 0.00046324669004803803,
      "loss": 5.4156,
      "mean_token_accuracy": 0.17032521963119507,
      "num_tokens": 43898649.0,
      "step": 21155
    },
    {
      "entropy": 5.807342195510865,
      "epoch": 1.9194484760522497,
      "grad_norm": 2.40625,
      "learning_rate": 0.000463228965361704,
      "loss": 5.5063,
      "mean_token_accuracy": 0.16349134743213653,
      "num_tokens": 43909369.0,
      "step": 21160
    },
    {
      "entropy": 5.734750127792358,
      "epoch": 1.919902031930334,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004632112367827469,
      "loss": 5.4171,
      "mean_token_accuracy": 0.17341679334640503,
      "num_tokens": 43919941.0,
      "step": 21165
    },
    {
      "entropy": 5.734735822677612,
      "epoch": 1.920355587808418,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004631935043115335,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1695522353053093,
      "num_tokens": 43929710.0,
      "step": 21170
    },
    {
      "entropy": 5.732624673843384,
      "epoch": 1.920809143686502,
      "grad_norm": 3.328125,
      "learning_rate": 0.00046317576794843056,
      "loss": 5.3896,
      "mean_token_accuracy": 0.1720599889755249,
      "num_tokens": 43939850.0,
      "step": 21175
    },
    {
      "entropy": 5.688083791732788,
      "epoch": 1.9212626995645863,
      "grad_norm": 4.5,
      "learning_rate": 0.00046315802769380474,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17686007171869278,
      "num_tokens": 43950067.0,
      "step": 21180
    },
    {
      "entropy": 5.667056083679199,
      "epoch": 1.9217162554426706,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004631402835480229,
      "loss": 5.3954,
      "mean_token_accuracy": 0.17611535936594008,
      "num_tokens": 43960029.0,
      "step": 21185
    },
    {
      "entropy": 5.731749725341797,
      "epoch": 1.9221698113207548,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046312253551145213,
      "loss": 5.4192,
      "mean_token_accuracy": 0.17075449079275132,
      "num_tokens": 43971221.0,
      "step": 21190
    },
    {
      "entropy": 5.731399154663086,
      "epoch": 1.9226233671988389,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046310478358445935,
      "loss": 5.439,
      "mean_token_accuracy": 0.16988783776760102,
      "num_tokens": 43981579.0,
      "step": 21195
    },
    {
      "entropy": 5.774648904800415,
      "epoch": 1.9230769230769231,
      "grad_norm": 2.125,
      "learning_rate": 0.0004630870277674117,
      "loss": 5.5309,
      "mean_token_accuracy": 0.1622469589114189,
      "num_tokens": 43991132.0,
      "step": 21200
    },
    {
      "entropy": 5.671022844314575,
      "epoch": 1.9235304789550072,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004630692680606764,
      "loss": 5.3315,
      "mean_token_accuracy": 0.17495701909065248,
      "num_tokens": 44001255.0,
      "step": 21205
    },
    {
      "entropy": 5.619173145294189,
      "epoch": 1.9239840348330914,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004630515044646207,
      "loss": 5.2913,
      "mean_token_accuracy": 0.17650680989027023,
      "num_tokens": 44011275.0,
      "step": 21210
    },
    {
      "entropy": 5.709741592407227,
      "epoch": 1.9244375907111757,
      "grad_norm": 2.03125,
      "learning_rate": 0.000463033736979612,
      "loss": 5.4417,
      "mean_token_accuracy": 0.1713992863893509,
      "num_tokens": 44022524.0,
      "step": 21215
    },
    {
      "entropy": 5.705796718597412,
      "epoch": 1.9248911465892597,
      "grad_norm": 3.5625,
      "learning_rate": 0.00046301596560601767,
      "loss": 5.3334,
      "mean_token_accuracy": 0.17179768830537795,
      "num_tokens": 44033955.0,
      "step": 21220
    },
    {
      "entropy": 5.695188093185425,
      "epoch": 1.925344702467344,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004629981903442052,
      "loss": 5.4724,
      "mean_token_accuracy": 0.1643437772989273,
      "num_tokens": 44045149.0,
      "step": 21225
    },
    {
      "entropy": 5.7154402256011965,
      "epoch": 1.925798258345428,
      "grad_norm": 3.5,
      "learning_rate": 0.0004629804111945424,
      "loss": 5.4375,
      "mean_token_accuracy": 0.16745451390743255,
      "num_tokens": 44055280.0,
      "step": 21230
    },
    {
      "entropy": 5.7545829772949215,
      "epoch": 1.9262518142235123,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004629626281573966,
      "loss": 5.4074,
      "mean_token_accuracy": 0.17609162479639054,
      "num_tokens": 44065578.0,
      "step": 21235
    },
    {
      "entropy": 5.685199356079101,
      "epoch": 1.9267053701015966,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004629448412331359,
      "loss": 5.3486,
      "mean_token_accuracy": 0.17474437803030013,
      "num_tokens": 44075454.0,
      "step": 21240
    },
    {
      "entropy": 5.716744089126587,
      "epoch": 1.9271589259796809,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004629270504221279,
      "loss": 5.4361,
      "mean_token_accuracy": 0.16866510808467866,
      "num_tokens": 44085680.0,
      "step": 21245
    },
    {
      "entropy": 5.6216064453125,
      "epoch": 1.927612481857765,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004629092557247407,
      "loss": 5.3467,
      "mean_token_accuracy": 0.17550325840711595,
      "num_tokens": 44097177.0,
      "step": 21250
    },
    {
      "entropy": 5.732069444656372,
      "epoch": 1.928066037735849,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004628914571413422,
      "loss": 5.4681,
      "mean_token_accuracy": 0.16826879531145095,
      "num_tokens": 44107251.0,
      "step": 21255
    },
    {
      "entropy": 5.718685293197632,
      "epoch": 1.9285195936139332,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046287365467230035,
      "loss": 5.4457,
      "mean_token_accuracy": 0.16722480952739716,
      "num_tokens": 44117943.0,
      "step": 21260
    },
    {
      "entropy": 5.778623867034912,
      "epoch": 1.9289731494920175,
      "grad_norm": 2.25,
      "learning_rate": 0.00046285584831798354,
      "loss": 5.4758,
      "mean_token_accuracy": 0.17338245064020158,
      "num_tokens": 44128608.0,
      "step": 21265
    },
    {
      "entropy": 5.73069577217102,
      "epoch": 1.9294267053701017,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046283803807875996,
      "loss": 5.428,
      "mean_token_accuracy": 0.1736005201935768,
      "num_tokens": 44138809.0,
      "step": 21270
    },
    {
      "entropy": 5.701419830322266,
      "epoch": 1.9298802612481858,
      "grad_norm": 3.65625,
      "learning_rate": 0.00046282022395499785,
      "loss": 5.427,
      "mean_token_accuracy": 0.1729208067059517,
      "num_tokens": 44148532.0,
      "step": 21275
    },
    {
      "entropy": 5.78540415763855,
      "epoch": 1.9303338171262698,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004628024059470656,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1680850312113762,
      "num_tokens": 44158628.0,
      "step": 21280
    },
    {
      "entropy": 5.67252402305603,
      "epoch": 1.930787373004354,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004627845840553318,
      "loss": 5.3172,
      "mean_token_accuracy": 0.1821717768907547,
      "num_tokens": 44168574.0,
      "step": 21285
    },
    {
      "entropy": 5.748472881317139,
      "epoch": 1.9312409288824384,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046276675828016493,
      "loss": 5.5195,
      "mean_token_accuracy": 0.170108462870121,
      "num_tokens": 44180207.0,
      "step": 21290
    },
    {
      "entropy": 5.789259195327759,
      "epoch": 1.9316944847605226,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004627489286219336,
      "loss": 5.5174,
      "mean_token_accuracy": 0.16419174373149872,
      "num_tokens": 44190473.0,
      "step": 21295
    },
    {
      "entropy": 5.797487688064575,
      "epoch": 1.9321480406386067,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004627310950810066,
      "loss": 5.4821,
      "mean_token_accuracy": 0.17172520756721496,
      "num_tokens": 44201150.0,
      "step": 21300
    },
    {
      "entropy": 5.690570878982544,
      "epoch": 1.9326015965166907,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004627132576577528,
      "loss": 5.4145,
      "mean_token_accuracy": 0.16568765342235564,
      "num_tokens": 44211272.0,
      "step": 21305
    },
    {
      "entropy": 5.7188849449157715,
      "epoch": 1.933055152394775,
      "grad_norm": 3.546875,
      "learning_rate": 0.000462695416352541,
      "loss": 5.4846,
      "mean_token_accuracy": 0.16645297259092331,
      "num_tokens": 44223556.0,
      "step": 21310
    },
    {
      "entropy": 5.738967514038086,
      "epoch": 1.9335087082728593,
      "grad_norm": 2.890625,
      "learning_rate": 0.00046267757116574,
      "loss": 5.3964,
      "mean_token_accuracy": 0.1765765890479088,
      "num_tokens": 44233911.0,
      "step": 21315
    },
    {
      "entropy": 5.74929986000061,
      "epoch": 1.9339622641509435,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004626597220977191,
      "loss": 5.4609,
      "mean_token_accuracy": 0.172481569647789,
      "num_tokens": 44244108.0,
      "step": 21320
    },
    {
      "entropy": 5.6462208271026615,
      "epoch": 1.9344158200290276,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046264186914884737,
      "loss": 5.3628,
      "mean_token_accuracy": 0.17447396069765092,
      "num_tokens": 44253604.0,
      "step": 21325
    },
    {
      "entropy": 5.691541004180908,
      "epoch": 1.9348693759071116,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046262401231949386,
      "loss": 5.4556,
      "mean_token_accuracy": 0.165654356777668,
      "num_tokens": 44263573.0,
      "step": 21330
    },
    {
      "entropy": 5.727162218093872,
      "epoch": 1.9353229317851959,
      "grad_norm": 2.921875,
      "learning_rate": 0.00046260615161002803,
      "loss": 5.4093,
      "mean_token_accuracy": 0.17006892561912537,
      "num_tokens": 44274071.0,
      "step": 21335
    },
    {
      "entropy": 5.818210220336914,
      "epoch": 1.9357764876632801,
      "grad_norm": 2.25,
      "learning_rate": 0.0004625882870208192,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16988101601600647,
      "num_tokens": 44283997.0,
      "step": 21340
    },
    {
      "entropy": 5.707959794998169,
      "epoch": 1.9362300435413644,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046257041855223674,
      "loss": 5.4131,
      "mean_token_accuracy": 0.17091355174779893,
      "num_tokens": 44295024.0,
      "step": 21345
    },
    {
      "entropy": 5.651451349258423,
      "epoch": 1.9366835994194485,
      "grad_norm": 2.375,
      "learning_rate": 0.00046255254620465033,
      "loss": 5.3565,
      "mean_token_accuracy": 0.17555786520242692,
      "num_tokens": 44305476.0,
      "step": 21350
    },
    {
      "entropy": 5.721776103973388,
      "epoch": 1.9371371552975325,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046253466997842936,
      "loss": 5.3534,
      "mean_token_accuracy": 0.1730489730834961,
      "num_tokens": 44316101.0,
      "step": 21355
    },
    {
      "entropy": 5.6711047172546385,
      "epoch": 1.9375907111756168,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004625167898739436,
      "loss": 5.2733,
      "mean_token_accuracy": 0.18071198016405104,
      "num_tokens": 44325058.0,
      "step": 21360
    },
    {
      "entropy": 5.818714189529419,
      "epoch": 1.938044267053701,
      "grad_norm": 2.15625,
      "learning_rate": 0.000462498905891563,
      "loss": 5.6217,
      "mean_token_accuracy": 0.15535377860069274,
      "num_tokens": 44336585.0,
      "step": 21365
    },
    {
      "entropy": 5.735881185531616,
      "epoch": 1.9384978229317853,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004624810180316571,
      "loss": 5.3992,
      "mean_token_accuracy": 0.1729408696293831,
      "num_tokens": 44346911.0,
      "step": 21370
    },
    {
      "entropy": 5.706629753112793,
      "epoch": 1.9389513788098693,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046246312629459596,
      "loss": 5.3834,
      "mean_token_accuracy": 0.1706651970744133,
      "num_tokens": 44358040.0,
      "step": 21375
    },
    {
      "entropy": 5.748434972763062,
      "epoch": 1.9394049346879536,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004624452306807497,
      "loss": 5.4146,
      "mean_token_accuracy": 0.17282870411872864,
      "num_tokens": 44368781.0,
      "step": 21380
    },
    {
      "entropy": 5.635067367553711,
      "epoch": 1.9398584905660377,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004624273311904882,
      "loss": 5.3564,
      "mean_token_accuracy": 0.17131560295820236,
      "num_tokens": 44378809.0,
      "step": 21385
    },
    {
      "entropy": 5.797924327850342,
      "epoch": 1.940312046444122,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046240942782418184,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16609262824058532,
      "num_tokens": 44388517.0,
      "step": 21390
    },
    {
      "entropy": 5.820730924606323,
      "epoch": 1.9407656023222062,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004623915205822007,
      "loss": 5.4653,
      "mean_token_accuracy": 0.16799347400665282,
      "num_tokens": 44399364.0,
      "step": 21395
    },
    {
      "entropy": 5.652327680587769,
      "epoch": 1.9412191582002902,
      "grad_norm": 2.125,
      "learning_rate": 0.00046237360946491514,
      "loss": 5.3836,
      "mean_token_accuracy": 0.16705659925937652,
      "num_tokens": 44409506.0,
      "step": 21400
    },
    {
      "entropy": 5.728401613235474,
      "epoch": 1.9416727140783745,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004623556944726956,
      "loss": 5.4276,
      "mean_token_accuracy": 0.17010619789361953,
      "num_tokens": 44419446.0,
      "step": 21405
    },
    {
      "entropy": 5.701863241195679,
      "epoch": 1.9421262699564585,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046233777560591256,
      "loss": 5.4156,
      "mean_token_accuracy": 0.1724273219704628,
      "num_tokens": 44429560.0,
      "step": 21410
    },
    {
      "entropy": 5.721386861801148,
      "epoch": 1.9425798258345428,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046231985286493654,
      "loss": 5.4024,
      "mean_token_accuracy": 0.16617671847343446,
      "num_tokens": 44439640.0,
      "step": 21415
    },
    {
      "entropy": 5.7651355266571045,
      "epoch": 1.943033381712627,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046230192625013836,
      "loss": 5.4348,
      "mean_token_accuracy": 0.1665117159485817,
      "num_tokens": 44450480.0,
      "step": 21420
    },
    {
      "entropy": 5.6800940990447994,
      "epoch": 1.9434869375907113,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046228399576188844,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17209070771932602,
      "num_tokens": 44460633.0,
      "step": 21425
    },
    {
      "entropy": 5.734530830383301,
      "epoch": 1.9439404934687954,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046226606140055784,
      "loss": 5.3855,
      "mean_token_accuracy": 0.17306210845708847,
      "num_tokens": 44470917.0,
      "step": 21430
    },
    {
      "entropy": 5.6989844799041744,
      "epoch": 1.9443940493468794,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046224812316651736,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16662354320287703,
      "num_tokens": 44482538.0,
      "step": 21435
    },
    {
      "entropy": 5.740198469161987,
      "epoch": 1.9448476052249637,
      "grad_norm": 3.109375,
      "learning_rate": 0.000462230181060138,
      "loss": 5.4445,
      "mean_token_accuracy": 0.16873575896024703,
      "num_tokens": 44493518.0,
      "step": 21440
    },
    {
      "entropy": 5.7327526092529295,
      "epoch": 1.945301161103048,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046221223508179077,
      "loss": 5.4347,
      "mean_token_accuracy": 0.1733684703707695,
      "num_tokens": 44503685.0,
      "step": 21445
    },
    {
      "entropy": 5.717018079757691,
      "epoch": 1.9457547169811322,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046219428523184685,
      "loss": 5.3905,
      "mean_token_accuracy": 0.17800275087356568,
      "num_tokens": 44514015.0,
      "step": 21450
    },
    {
      "entropy": 5.713047313690185,
      "epoch": 1.9462082728592163,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004621763315106774,
      "loss": 5.4251,
      "mean_token_accuracy": 0.1718946024775505,
      "num_tokens": 44525187.0,
      "step": 21455
    },
    {
      "entropy": 5.633773851394653,
      "epoch": 1.9466618287373003,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046215837391865364,
      "loss": 5.3935,
      "mean_token_accuracy": 0.176674285531044,
      "num_tokens": 44535964.0,
      "step": 21460
    },
    {
      "entropy": 5.61548867225647,
      "epoch": 1.9471153846153846,
      "grad_norm": 4.53125,
      "learning_rate": 0.00046214041245614713,
      "loss": 5.3699,
      "mean_token_accuracy": 0.16918833255767823,
      "num_tokens": 44546985.0,
      "step": 21465
    },
    {
      "entropy": 5.715643978118896,
      "epoch": 1.9475689404934688,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004621224471235291,
      "loss": 5.4352,
      "mean_token_accuracy": 0.17264892160892487,
      "num_tokens": 44556888.0,
      "step": 21470
    },
    {
      "entropy": 5.683076477050781,
      "epoch": 1.9480224963715531,
      "grad_norm": 2.5,
      "learning_rate": 0.00046210447792117114,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1696619436144829,
      "num_tokens": 44566857.0,
      "step": 21475
    },
    {
      "entropy": 5.743082857131958,
      "epoch": 1.9484760522496372,
      "grad_norm": 2.75,
      "learning_rate": 0.000462086504849445,
      "loss": 5.4424,
      "mean_token_accuracy": 0.16844300031661988,
      "num_tokens": 44578152.0,
      "step": 21480
    },
    {
      "entropy": 5.757699966430664,
      "epoch": 1.9489296081277212,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004620685279087222,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1756268098950386,
      "num_tokens": 44588814.0,
      "step": 21485
    },
    {
      "entropy": 5.709230279922485,
      "epoch": 1.9493831640058055,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046205054709937474,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17073371708393098,
      "num_tokens": 44599366.0,
      "step": 21490
    },
    {
      "entropy": 5.757882308959961,
      "epoch": 1.9498367198838897,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004620325624217741,
      "loss": 5.5019,
      "mean_token_accuracy": 0.16665035337209702,
      "num_tokens": 44609962.0,
      "step": 21495
    },
    {
      "entropy": 5.796237659454346,
      "epoch": 1.950290275761974,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004620145738762925,
      "loss": 5.4418,
      "mean_token_accuracy": 0.17539315521717072,
      "num_tokens": 44619917.0,
      "step": 21500
    },
    {
      "entropy": 5.628072023391724,
      "epoch": 1.950743831640058,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004619965814633019,
      "loss": 5.4107,
      "mean_token_accuracy": 0.1744246929883957,
      "num_tokens": 44629723.0,
      "step": 21505
    },
    {
      "entropy": 5.677640771865844,
      "epoch": 1.951197387518142,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004619785851831742,
      "loss": 5.4438,
      "mean_token_accuracy": 0.17118796110153198,
      "num_tokens": 44641011.0,
      "step": 21510
    },
    {
      "entropy": 5.749798440933228,
      "epoch": 1.9516509433962264,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046196058503628176,
      "loss": 5.3674,
      "mean_token_accuracy": 0.17646342068910598,
      "num_tokens": 44651154.0,
      "step": 21515
    },
    {
      "entropy": 5.726172256469726,
      "epoch": 1.9521044992743106,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004619425810229968,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16879691779613495,
      "num_tokens": 44661672.0,
      "step": 21520
    },
    {
      "entropy": 5.668769598007202,
      "epoch": 1.952558055152395,
      "grad_norm": 3.375,
      "learning_rate": 0.0004619245731436917,
      "loss": 5.348,
      "mean_token_accuracy": 0.17809549123048782,
      "num_tokens": 44671449.0,
      "step": 21525
    },
    {
      "entropy": 5.729270792007446,
      "epoch": 1.953011611030479,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046190656139873863,
      "loss": 5.3678,
      "mean_token_accuracy": 0.17476011514663697,
      "num_tokens": 44681977.0,
      "step": 21530
    },
    {
      "entropy": 5.666712045669556,
      "epoch": 1.953465166908563,
      "grad_norm": 17.375,
      "learning_rate": 0.0004618885457885103,
      "loss": 5.3764,
      "mean_token_accuracy": 0.1759292870759964,
      "num_tokens": 44691946.0,
      "step": 21535
    },
    {
      "entropy": 5.657412385940551,
      "epoch": 1.9539187227866472,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004618705263133793,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17060646563768386,
      "num_tokens": 44702012.0,
      "step": 21540
    },
    {
      "entropy": 5.693317699432373,
      "epoch": 1.9543722786647315,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004618525029737182,
      "loss": 5.3576,
      "mean_token_accuracy": 0.17610587626695634,
      "num_tokens": 44711605.0,
      "step": 21545
    },
    {
      "entropy": 5.782636642456055,
      "epoch": 1.9548258345428158,
      "grad_norm": 3.53125,
      "learning_rate": 0.00046183447576989966,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1759013742208481,
      "num_tokens": 44722765.0,
      "step": 21550
    },
    {
      "entropy": 5.7643180847167965,
      "epoch": 1.9552793904208998,
      "grad_norm": 7.15625,
      "learning_rate": 0.00046181644470229656,
      "loss": 5.3986,
      "mean_token_accuracy": 0.172443687915802,
      "num_tokens": 44732082.0,
      "step": 21555
    },
    {
      "entropy": 5.660933208465576,
      "epoch": 1.955732946298984,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004617984097712817,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17747943103313446,
      "num_tokens": 44742054.0,
      "step": 21560
    },
    {
      "entropy": 5.661222505569458,
      "epoch": 1.9561865021770681,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004617803709772282,
      "loss": 5.4431,
      "mean_token_accuracy": 0.17361314296722413,
      "num_tokens": 44752954.0,
      "step": 21565
    },
    {
      "entropy": 5.718326616287231,
      "epoch": 1.9566400580551524,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004617623283205091,
      "loss": 5.4857,
      "mean_token_accuracy": 0.1693657234311104,
      "num_tokens": 44764277.0,
      "step": 21570
    },
    {
      "entropy": 5.753034925460815,
      "epoch": 1.9570936139332367,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046174428180149733,
      "loss": 5.4548,
      "mean_token_accuracy": 0.17381251007318496,
      "num_tokens": 44775429.0,
      "step": 21575
    },
    {
      "entropy": 5.775501918792725,
      "epoch": 1.9575471698113207,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004617262314205663,
      "loss": 5.4553,
      "mean_token_accuracy": 0.16530355662107468,
      "num_tokens": 44786390.0,
      "step": 21580
    },
    {
      "entropy": 5.732374143600464,
      "epoch": 1.958000725689405,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004617081771780893,
      "loss": 5.3998,
      "mean_token_accuracy": 0.17692067623138427,
      "num_tokens": 44796473.0,
      "step": 21585
    },
    {
      "entropy": 5.686982536315918,
      "epoch": 1.958454281567489,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046169011907443944,
      "loss": 5.3636,
      "mean_token_accuracy": 0.17411493510007858,
      "num_tokens": 44806379.0,
      "step": 21590
    },
    {
      "entropy": 5.6773412227630615,
      "epoch": 1.9589078374455733,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004616720571099904,
      "loss": 5.4588,
      "mean_token_accuracy": 0.16780052185058594,
      "num_tokens": 44817659.0,
      "step": 21595
    },
    {
      "entropy": 5.708354425430298,
      "epoch": 1.9593613933236576,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004616539912851157,
      "loss": 5.3461,
      "mean_token_accuracy": 0.17052143216133117,
      "num_tokens": 44827073.0,
      "step": 21600
    },
    {
      "entropy": 5.671363735198975,
      "epoch": 1.9598149492017418,
      "grad_norm": 2.125,
      "learning_rate": 0.00046163592160018886,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17898421734571457,
      "num_tokens": 44837019.0,
      "step": 21605
    },
    {
      "entropy": 5.651212120056153,
      "epoch": 1.9602685050798259,
      "grad_norm": 2.875,
      "learning_rate": 0.0004616178480555836,
      "loss": 5.3856,
      "mean_token_accuracy": 0.17942943871021272,
      "num_tokens": 44846670.0,
      "step": 21610
    },
    {
      "entropy": 5.627276945114136,
      "epoch": 1.96072206095791,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004615997706516737,
      "loss": 5.3365,
      "mean_token_accuracy": 0.17602232247591018,
      "num_tokens": 44857016.0,
      "step": 21615
    },
    {
      "entropy": 5.770315885543823,
      "epoch": 1.9611756168359942,
      "grad_norm": 2.265625,
      "learning_rate": 0.000461581689388833,
      "loss": 5.5041,
      "mean_token_accuracy": 0.161636783182621,
      "num_tokens": 44868180.0,
      "step": 21620
    },
    {
      "entropy": 5.711691236495971,
      "epoch": 1.9616291727140784,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046156360426743536,
      "loss": 5.3287,
      "mean_token_accuracy": 0.1762249708175659,
      "num_tokens": 44877864.0,
      "step": 21625
    },
    {
      "entropy": 5.655797290802002,
      "epoch": 1.9620827285921627,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004615455152878549,
      "loss": 5.3925,
      "mean_token_accuracy": 0.17761609852313995,
      "num_tokens": 44887737.0,
      "step": 21630
    },
    {
      "entropy": 5.6262890815734865,
      "epoch": 1.9625362844702468,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004615274224504657,
      "loss": 5.4183,
      "mean_token_accuracy": 0.16742431819438935,
      "num_tokens": 44898669.0,
      "step": 21635
    },
    {
      "entropy": 5.754499101638794,
      "epoch": 1.9629898403483308,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046150932575564173,
      "loss": 5.4819,
      "mean_token_accuracy": 0.17275461107492446,
      "num_tokens": 44909074.0,
      "step": 21640
    },
    {
      "entropy": 5.729836416244507,
      "epoch": 1.963443396226415,
      "grad_norm": 3.515625,
      "learning_rate": 0.00046149122520375746,
      "loss": 5.3663,
      "mean_token_accuracy": 0.17171652019023895,
      "num_tokens": 44919468.0,
      "step": 21645
    },
    {
      "entropy": 5.706184053421021,
      "epoch": 1.9638969521044993,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004614731207951871,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1659515157341957,
      "num_tokens": 44930470.0,
      "step": 21650
    },
    {
      "entropy": 5.6447625160217285,
      "epoch": 1.9643505079825836,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004614550125303051,
      "loss": 5.4535,
      "mean_token_accuracy": 0.17559978514909744,
      "num_tokens": 44940611.0,
      "step": 21655
    },
    {
      "entropy": 5.743188190460205,
      "epoch": 1.9648040638606676,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004614369004094859,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1725083738565445,
      "num_tokens": 44950985.0,
      "step": 21660
    },
    {
      "entropy": 5.744107294082641,
      "epoch": 1.9652576197387517,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046141878443310415,
      "loss": 5.4629,
      "mean_token_accuracy": 0.1678045153617859,
      "num_tokens": 44961858.0,
      "step": 21665
    },
    {
      "entropy": 5.712519121170044,
      "epoch": 1.965711175616836,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00046140066460153444,
      "loss": 5.4248,
      "mean_token_accuracy": 0.1698423981666565,
      "num_tokens": 44972048.0,
      "step": 21670
    },
    {
      "entropy": 5.69265513420105,
      "epoch": 1.9661647314949202,
      "grad_norm": 5.125,
      "learning_rate": 0.00046138254091515136,
      "loss": 5.3582,
      "mean_token_accuracy": 0.17209556251764296,
      "num_tokens": 44981835.0,
      "step": 21675
    },
    {
      "entropy": 5.722651481628418,
      "epoch": 1.9666182873730045,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046136441337432996,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1783606693148613,
      "num_tokens": 44992108.0,
      "step": 21680
    },
    {
      "entropy": 5.840202379226684,
      "epoch": 1.9670718432510885,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046134628197944496,
      "loss": 5.4732,
      "mean_token_accuracy": 0.1661076009273529,
      "num_tokens": 45002105.0,
      "step": 21685
    },
    {
      "entropy": 5.695396947860718,
      "epoch": 1.9675253991291726,
      "grad_norm": 2.375,
      "learning_rate": 0.00046132814673087143,
      "loss": 5.4359,
      "mean_token_accuracy": 0.16680711060762404,
      "num_tokens": 45013934.0,
      "step": 21690
    },
    {
      "entropy": 5.655059003829956,
      "epoch": 1.9679789550072568,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004613100076289842,
      "loss": 5.4026,
      "mean_token_accuracy": 0.17214062213897705,
      "num_tokens": 45023851.0,
      "step": 21695
    },
    {
      "entropy": 5.662230777740478,
      "epoch": 1.968432510885341,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004612918646741586,
      "loss": 5.3216,
      "mean_token_accuracy": 0.17144415229558946,
      "num_tokens": 45034140.0,
      "step": 21700
    },
    {
      "entropy": 5.664430856704712,
      "epoch": 1.9688860667634254,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004612737178667698,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1724102258682251,
      "num_tokens": 45044452.0,
      "step": 21705
    },
    {
      "entropy": 5.682538175582886,
      "epoch": 1.9693396226415094,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046125556720719304,
      "loss": 5.4031,
      "mean_token_accuracy": 0.17221750169992447,
      "num_tokens": 45053845.0,
      "step": 21710
    },
    {
      "entropy": 5.693418550491333,
      "epoch": 1.9697931785195935,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004612374126958037,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17644451111555098,
      "num_tokens": 45063973.0,
      "step": 21715
    },
    {
      "entropy": 5.781726932525634,
      "epoch": 1.9702467343976777,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004612192543329771,
      "loss": 5.5205,
      "mean_token_accuracy": 0.16642228364944459,
      "num_tokens": 45074879.0,
      "step": 21720
    },
    {
      "entropy": 5.697332715988159,
      "epoch": 1.970700290275762,
      "grad_norm": 2.59375,
      "learning_rate": 0.000461201092119089,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17980997562408446,
      "num_tokens": 45085161.0,
      "step": 21725
    },
    {
      "entropy": 5.702481269836426,
      "epoch": 1.9711538461538463,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004611829260545148,
      "loss": 5.4574,
      "mean_token_accuracy": 0.16656685322523118,
      "num_tokens": 45095862.0,
      "step": 21730
    },
    {
      "entropy": 5.746847295761109,
      "epoch": 1.9716074020319303,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046116475613963026,
      "loss": 5.402,
      "mean_token_accuracy": 0.17104933112859727,
      "num_tokens": 45106396.0,
      "step": 21735
    },
    {
      "entropy": 5.765971374511719,
      "epoch": 1.9720609579100146,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046114658237481114,
      "loss": 5.3396,
      "mean_token_accuracy": 0.17539280652999878,
      "num_tokens": 45117178.0,
      "step": 21740
    },
    {
      "entropy": 5.580171251296997,
      "epoch": 1.9725145137880986,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004611284047604332,
      "loss": 5.3255,
      "mean_token_accuracy": 0.18202819675207138,
      "num_tokens": 45126571.0,
      "step": 21745
    },
    {
      "entropy": 5.721501636505127,
      "epoch": 1.9729680696661829,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046111022329687246,
      "loss": 5.4601,
      "mean_token_accuracy": 0.1664485976099968,
      "num_tokens": 45137247.0,
      "step": 21750
    },
    {
      "entropy": 5.8217936038970945,
      "epoch": 1.9734216255442671,
      "grad_norm": 2.625,
      "learning_rate": 0.00046109203798450483,
      "loss": 5.403,
      "mean_token_accuracy": 0.1694312497973442,
      "num_tokens": 45147531.0,
      "step": 21755
    },
    {
      "entropy": 5.69701886177063,
      "epoch": 1.9738751814223512,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004610738488237065,
      "loss": 5.384,
      "mean_token_accuracy": 0.17849550247192383,
      "num_tokens": 45158312.0,
      "step": 21760
    },
    {
      "entropy": 5.675541496276855,
      "epoch": 1.9743287373004355,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004610556558148535,
      "loss": 5.457,
      "mean_token_accuracy": 0.17263800650835037,
      "num_tokens": 45169754.0,
      "step": 21765
    },
    {
      "entropy": 5.701108932495117,
      "epoch": 1.9747822931785195,
      "grad_norm": 2.25,
      "learning_rate": 0.0004610374589583222,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1692351520061493,
      "num_tokens": 45180428.0,
      "step": 21770
    },
    {
      "entropy": 5.7201886653900145,
      "epoch": 1.9752358490566038,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004610192582544888,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17284373790025712,
      "num_tokens": 45191558.0,
      "step": 21775
    },
    {
      "entropy": 5.6706493377685545,
      "epoch": 1.975689404934688,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004610010537037297,
      "loss": 5.3502,
      "mean_token_accuracy": 0.1720207929611206,
      "num_tokens": 45201567.0,
      "step": 21780
    },
    {
      "entropy": 5.74292950630188,
      "epoch": 1.9761429608127723,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004609828453064214,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1646181359887123,
      "num_tokens": 45211359.0,
      "step": 21785
    },
    {
      "entropy": 5.766177368164063,
      "epoch": 1.9765965166908563,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046096463306294057,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1698679506778717,
      "num_tokens": 45222106.0,
      "step": 21790
    },
    {
      "entropy": 5.692195034027099,
      "epoch": 1.9770500725689404,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046094641697366366,
      "loss": 5.4379,
      "mean_token_accuracy": 0.1649845287203789,
      "num_tokens": 45233169.0,
      "step": 21795
    },
    {
      "entropy": 5.736575508117676,
      "epoch": 1.9775036284470247,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004609281970389675,
      "loss": 5.4855,
      "mean_token_accuracy": 0.16727641522884368,
      "num_tokens": 45243627.0,
      "step": 21800
    },
    {
      "entropy": 5.786532545089722,
      "epoch": 1.977957184325109,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046090997325922887,
      "loss": 5.3771,
      "mean_token_accuracy": 0.1760561615228653,
      "num_tokens": 45253730.0,
      "step": 21805
    },
    {
      "entropy": 5.7310832977294925,
      "epoch": 1.9784107402031932,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046089174563482456,
      "loss": 5.4779,
      "mean_token_accuracy": 0.17503363341093064,
      "num_tokens": 45263443.0,
      "step": 21810
    },
    {
      "entropy": 5.695977020263672,
      "epoch": 1.9788642960812772,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004608735141661316,
      "loss": 5.4012,
      "mean_token_accuracy": 0.1744118094444275,
      "num_tokens": 45275298.0,
      "step": 21815
    },
    {
      "entropy": 5.72073016166687,
      "epoch": 1.9793178519593613,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046085527885352694,
      "loss": 5.4298,
      "mean_token_accuracy": 0.1724221646785736,
      "num_tokens": 45285628.0,
      "step": 21820
    },
    {
      "entropy": 5.657871675491333,
      "epoch": 1.9797714078374455,
      "grad_norm": 1.625,
      "learning_rate": 0.0004608370396973878,
      "loss": 5.3468,
      "mean_token_accuracy": 0.17843465656042098,
      "num_tokens": 45295559.0,
      "step": 21825
    },
    {
      "entropy": 5.680355405807495,
      "epoch": 1.9802249637155298,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046081879669809133,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17344353646039962,
      "num_tokens": 45305395.0,
      "step": 21830
    },
    {
      "entropy": 5.691047716140747,
      "epoch": 1.980678519593614,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004608005498560148,
      "loss": 5.3994,
      "mean_token_accuracy": 0.17746995240449906,
      "num_tokens": 45316763.0,
      "step": 21835
    },
    {
      "entropy": 5.772811841964722,
      "epoch": 1.9811320754716981,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004607822991715356,
      "loss": 5.5145,
      "mean_token_accuracy": 0.1682760089635849,
      "num_tokens": 45326484.0,
      "step": 21840
    },
    {
      "entropy": 5.748633623123169,
      "epoch": 1.9815856313497822,
      "grad_norm": 1.828125,
      "learning_rate": 0.000460764044645031,
      "loss": 5.4436,
      "mean_token_accuracy": 0.16930464804172515,
      "num_tokens": 45335792.0,
      "step": 21845
    },
    {
      "entropy": 5.734651851654053,
      "epoch": 1.9820391872278664,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004607457862768787,
      "loss": 5.3717,
      "mean_token_accuracy": 0.16768748313188553,
      "num_tokens": 45347247.0,
      "step": 21850
    },
    {
      "entropy": 5.758577060699463,
      "epoch": 1.9824927431059507,
      "grad_norm": 5.65625,
      "learning_rate": 0.0004607275240674562,
      "loss": 5.4963,
      "mean_token_accuracy": 0.16803329288959504,
      "num_tokens": 45357994.0,
      "step": 21855
    },
    {
      "entropy": 5.645494651794434,
      "epoch": 1.982946298984035,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004607092580171411,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17490310966968536,
      "num_tokens": 45368170.0,
      "step": 21860
    },
    {
      "entropy": 5.762082481384278,
      "epoch": 1.983399854862119,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004606909881263113,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16980123221874238,
      "num_tokens": 45378696.0,
      "step": 21865
    },
    {
      "entropy": 5.726118803024292,
      "epoch": 1.983853410740203,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046067271439534445,
      "loss": 5.4677,
      "mean_token_accuracy": 0.17053177505731582,
      "num_tokens": 45388578.0,
      "step": 21870
    },
    {
      "entropy": 5.706598520278931,
      "epoch": 1.9843069666182873,
      "grad_norm": 1.75,
      "learning_rate": 0.00046065443682461864,
      "loss": 5.4629,
      "mean_token_accuracy": 0.17340354174375533,
      "num_tokens": 45398835.0,
      "step": 21875
    },
    {
      "entropy": 5.764643335342408,
      "epoch": 1.9847605224963716,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046063615541451177,
      "loss": 5.4321,
      "mean_token_accuracy": 0.17703914046287536,
      "num_tokens": 45409032.0,
      "step": 21880
    },
    {
      "entropy": 5.832647895812988,
      "epoch": 1.9852140783744558,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046061787016540184,
      "loss": 5.5794,
      "mean_token_accuracy": 0.16456862539052963,
      "num_tokens": 45420661.0,
      "step": 21885
    },
    {
      "entropy": 5.7013660907745365,
      "epoch": 1.98566763425254,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046059958107766706,
      "loss": 5.2833,
      "mean_token_accuracy": 0.184004744887352,
      "num_tokens": 45429751.0,
      "step": 21890
    },
    {
      "entropy": 5.698215627670288,
      "epoch": 1.986121190130624,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004605812881516857,
      "loss": 5.4329,
      "mean_token_accuracy": 0.17481494545936585,
      "num_tokens": 45439551.0,
      "step": 21895
    },
    {
      "entropy": 5.600321054458618,
      "epoch": 1.9865747460087082,
      "grad_norm": 1.5,
      "learning_rate": 0.00046056299138783595,
      "loss": 5.3165,
      "mean_token_accuracy": 0.18687519133090974,
      "num_tokens": 45449543.0,
      "step": 21900
    },
    {
      "entropy": 5.749264049530029,
      "epoch": 1.9870283018867925,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004605446907864963,
      "loss": 5.4145,
      "mean_token_accuracy": 0.17551022171974182,
      "num_tokens": 45459069.0,
      "step": 21905
    },
    {
      "entropy": 5.694426202774048,
      "epoch": 1.9874818577648767,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004605263863480452,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1662349820137024,
      "num_tokens": 45468987.0,
      "step": 21910
    },
    {
      "entropy": 5.703587007522583,
      "epoch": 1.9879354136429608,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046050807807286105,
      "loss": 5.4412,
      "mean_token_accuracy": 0.1712467700242996,
      "num_tokens": 45479781.0,
      "step": 21915
    },
    {
      "entropy": 5.742800855636597,
      "epoch": 1.988388969521045,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046048976596132265,
      "loss": 5.5475,
      "mean_token_accuracy": 0.15657341331243516,
      "num_tokens": 45490318.0,
      "step": 21920
    },
    {
      "entropy": 5.743636178970337,
      "epoch": 1.988842525399129,
      "grad_norm": 2.75,
      "learning_rate": 0.00046047145001380856,
      "loss": 5.3987,
      "mean_token_accuracy": 0.17373684346675872,
      "num_tokens": 45501514.0,
      "step": 21925
    },
    {
      "entropy": 5.749484920501709,
      "epoch": 1.9892960812772134,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046045313023069764,
      "loss": 5.3949,
      "mean_token_accuracy": 0.17546605914831162,
      "num_tokens": 45511727.0,
      "step": 21930
    },
    {
      "entropy": 5.752577686309815,
      "epoch": 1.9897496371552976,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004604348066123688,
      "loss": 5.4975,
      "mean_token_accuracy": 0.16670543402433396,
      "num_tokens": 45521807.0,
      "step": 21935
    },
    {
      "entropy": 5.72134313583374,
      "epoch": 1.9902031930333817,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004604164791592008,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1648574322462082,
      "num_tokens": 45532033.0,
      "step": 21940
    },
    {
      "entropy": 5.714197111129761,
      "epoch": 1.990656748911466,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046039814787157283,
      "loss": 5.4438,
      "mean_token_accuracy": 0.16579942852258683,
      "num_tokens": 45542806.0,
      "step": 21945
    },
    {
      "entropy": 5.7320143699646,
      "epoch": 1.99111030478955,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004603798127498639,
      "loss": 5.3213,
      "mean_token_accuracy": 0.18557538837194443,
      "num_tokens": 45551982.0,
      "step": 21950
    },
    {
      "entropy": 5.727459049224853,
      "epoch": 1.9915638606676342,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004603614737944532,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1699257269501686,
      "num_tokens": 45561998.0,
      "step": 21955
    },
    {
      "entropy": 5.667831754684448,
      "epoch": 1.9920174165457185,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046034313100571994,
      "loss": 5.4304,
      "mean_token_accuracy": 0.1728160098195076,
      "num_tokens": 45572246.0,
      "step": 21960
    },
    {
      "entropy": 5.68893723487854,
      "epoch": 1.9924709724238028,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004603247843840436,
      "loss": 5.4339,
      "mean_token_accuracy": 0.17423621863126754,
      "num_tokens": 45582251.0,
      "step": 21965
    },
    {
      "entropy": 5.689378356933593,
      "epoch": 1.9929245283018868,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004603064339298034,
      "loss": 5.3192,
      "mean_token_accuracy": 0.17483141124248505,
      "num_tokens": 45592749.0,
      "step": 21970
    },
    {
      "entropy": 5.6776899814605715,
      "epoch": 1.9933780841799709,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004602880796433789,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17640600502490997,
      "num_tokens": 45602412.0,
      "step": 21975
    },
    {
      "entropy": 5.7261169910430905,
      "epoch": 1.9938316400580551,
      "grad_norm": 2.890625,
      "learning_rate": 0.00046026972152514976,
      "loss": 5.416,
      "mean_token_accuracy": 0.17491375505924225,
      "num_tokens": 45612736.0,
      "step": 21980
    },
    {
      "entropy": 5.676369571685791,
      "epoch": 1.9942851959361394,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004602513595754956,
      "loss": 5.428,
      "mean_token_accuracy": 0.17092739939689636,
      "num_tokens": 45622952.0,
      "step": 21985
    },
    {
      "entropy": 5.7158019065856935,
      "epoch": 1.9947387518142237,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046023299379479606,
      "loss": 5.4101,
      "mean_token_accuracy": 0.17207694351673125,
      "num_tokens": 45633569.0,
      "step": 21990
    },
    {
      "entropy": 5.684228849411011,
      "epoch": 1.9951923076923077,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046021462418343104,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17707901895046235,
      "num_tokens": 45644347.0,
      "step": 21995
    },
    {
      "entropy": 5.7621660232543945,
      "epoch": 1.9956458635703918,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004601962507417804,
      "loss": 5.4571,
      "mean_token_accuracy": 0.1748166263103485,
      "num_tokens": 45655268.0,
      "step": 22000
    },
    {
      "entropy": 5.728942823410034,
      "epoch": 1.996099419448476,
      "grad_norm": 3.921875,
      "learning_rate": 0.00046017787347022403,
      "loss": 5.3642,
      "mean_token_accuracy": 0.17395815253257751,
      "num_tokens": 45664900.0,
      "step": 22005
    },
    {
      "entropy": 5.710017204284668,
      "epoch": 1.9965529753265603,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046015949236914217,
      "loss": 5.4708,
      "mean_token_accuracy": 0.17228570878505706,
      "num_tokens": 45675252.0,
      "step": 22010
    },
    {
      "entropy": 5.798247909545898,
      "epoch": 1.9970065312046446,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004601411074389147,
      "loss": 5.3702,
      "mean_token_accuracy": 0.18272842913866044,
      "num_tokens": 45686027.0,
      "step": 22015
    },
    {
      "entropy": 5.7311845302581785,
      "epoch": 1.9974600870827286,
      "grad_norm": 2.78125,
      "learning_rate": 0.000460122718679922,
      "loss": 5.3875,
      "mean_token_accuracy": 0.17537824362516402,
      "num_tokens": 45696436.0,
      "step": 22020
    },
    {
      "entropy": 5.732070589065552,
      "epoch": 1.9979136429608126,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004601043260925443,
      "loss": 5.4886,
      "mean_token_accuracy": 0.17846929728984834,
      "num_tokens": 45706509.0,
      "step": 22025
    },
    {
      "entropy": 5.770925807952881,
      "epoch": 1.998367198838897,
      "grad_norm": 3.140625,
      "learning_rate": 0.00046008592967716197,
      "loss": 5.5285,
      "mean_token_accuracy": 0.16454817354679108,
      "num_tokens": 45717933.0,
      "step": 22030
    },
    {
      "entropy": 5.709897041320801,
      "epoch": 1.9988207547169812,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004600675294341554,
      "loss": 5.4504,
      "mean_token_accuracy": 0.16305852681398392,
      "num_tokens": 45727966.0,
      "step": 22035
    },
    {
      "entropy": 5.715452814102173,
      "epoch": 1.9992743105950654,
      "grad_norm": 2.25,
      "learning_rate": 0.00046004912536390526,
      "loss": 5.4126,
      "mean_token_accuracy": 0.16885788589715958,
      "num_tokens": 45737649.0,
      "step": 22040
    },
    {
      "entropy": 5.718445539474487,
      "epoch": 1.9997278664731495,
      "grad_norm": 2.1875,
      "learning_rate": 0.000460030717466792,
      "loss": 5.3835,
      "mean_token_accuracy": 0.17012977600097656,
      "num_tokens": 45747561.0,
      "step": 22045
    },
    {
      "entropy": 5.688719844818115,
      "epoch": 2.0001814223512335,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004600123057431963,
      "loss": 5.369,
      "mean_token_accuracy": 0.17176604121923447,
      "num_tokens": 45758183.0,
      "step": 22050
    },
    {
      "entropy": 5.701321125030518,
      "epoch": 2.000634978229318,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000459993890193499,
      "loss": 5.3151,
      "mean_token_accuracy": 0.18433785438537598,
      "num_tokens": 45768006.0,
      "step": 22055
    },
    {
      "entropy": 5.785220336914063,
      "epoch": 2.001088534107402,
      "grad_norm": 5.84375,
      "learning_rate": 0.0004599754708180809,
      "loss": 5.4483,
      "mean_token_accuracy": 0.16705699414014816,
      "num_tokens": 45777981.0,
      "step": 22060
    },
    {
      "entropy": 5.6469463348388675,
      "epoch": 2.0015420899854863,
      "grad_norm": 3.5,
      "learning_rate": 0.000459957047617323,
      "loss": 5.3932,
      "mean_token_accuracy": 0.16517809629440308,
      "num_tokens": 45789083.0,
      "step": 22065
    },
    {
      "entropy": 5.680325078964233,
      "epoch": 2.0019956458635706,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045993862059160617,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17774357199668883,
      "num_tokens": 45799750.0,
      "step": 22070
    },
    {
      "entropy": 5.684410667419433,
      "epoch": 2.0024492017416544,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004599201897413116,
      "loss": 5.3301,
      "mean_token_accuracy": 0.17300645112991334,
      "num_tokens": 45810321.0,
      "step": 22075
    },
    {
      "entropy": 5.691515159606934,
      "epoch": 2.0029027576197387,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045990175506682024,
      "loss": 5.3385,
      "mean_token_accuracy": 0.18587927520275116,
      "num_tokens": 45820581.0,
      "step": 22080
    },
    {
      "entropy": 5.647567939758301,
      "epoch": 2.003356313497823,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045988331656851355,
      "loss": 5.3984,
      "mean_token_accuracy": 0.17206306159496307,
      "num_tokens": 45831504.0,
      "step": 22085
    },
    {
      "entropy": 5.703463792800903,
      "epoch": 2.003809869375907,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045986487424677284,
      "loss": 5.4283,
      "mean_token_accuracy": 0.17073594480752946,
      "num_tokens": 45841100.0,
      "step": 22090
    },
    {
      "entropy": 5.7728416442871096,
      "epoch": 2.0042634252539915,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004598464281019793,
      "loss": 5.4021,
      "mean_token_accuracy": 0.1762344092130661,
      "num_tokens": 45851647.0,
      "step": 22095
    },
    {
      "entropy": 5.72019853591919,
      "epoch": 2.0047169811320753,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045982797813451463,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17231757044792176,
      "num_tokens": 45862995.0,
      "step": 22100
    },
    {
      "entropy": 5.6790060043334964,
      "epoch": 2.0051705370101596,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045980952434476014,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17563431859016418,
      "num_tokens": 45872360.0,
      "step": 22105
    },
    {
      "entropy": 5.665389490127564,
      "epoch": 2.005624092888244,
      "grad_norm": 3.359375,
      "learning_rate": 0.00045979106673309765,
      "loss": 5.4155,
      "mean_token_accuracy": 0.16848233640193938,
      "num_tokens": 45883440.0,
      "step": 22110
    },
    {
      "entropy": 5.652190494537353,
      "epoch": 2.006077648766328,
      "grad_norm": 1.75,
      "learning_rate": 0.00045977260529990883,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17184120565652847,
      "num_tokens": 45894323.0,
      "step": 22115
    },
    {
      "entropy": 5.730787038803101,
      "epoch": 2.0065312046444124,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004597541400455755,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1712713584303856,
      "num_tokens": 45904544.0,
      "step": 22120
    },
    {
      "entropy": 5.763525009155273,
      "epoch": 2.006984760522496,
      "grad_norm": 1.75,
      "learning_rate": 0.00045973567097047933,
      "loss": 5.2934,
      "mean_token_accuracy": 0.1893812596797943,
      "num_tokens": 45913225.0,
      "step": 22125
    },
    {
      "entropy": 5.637686109542846,
      "epoch": 2.0074383164005805,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045971719807500253,
      "loss": 5.4141,
      "mean_token_accuracy": 0.17194016128778458,
      "num_tokens": 45923306.0,
      "step": 22130
    },
    {
      "entropy": 5.674941301345825,
      "epoch": 2.0078918722786647,
      "grad_norm": 5.375,
      "learning_rate": 0.000459698721359527,
      "loss": 5.3689,
      "mean_token_accuracy": 0.1757521316409111,
      "num_tokens": 45933243.0,
      "step": 22135
    },
    {
      "entropy": 5.682531452178955,
      "epoch": 2.008345428156749,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004596802408244348,
      "loss": 5.4174,
      "mean_token_accuracy": 0.17457915097475052,
      "num_tokens": 45944498.0,
      "step": 22140
    },
    {
      "entropy": 5.6345093727111815,
      "epoch": 2.0087989840348333,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004596617564701081,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17718230038881302,
      "num_tokens": 45954550.0,
      "step": 22145
    },
    {
      "entropy": 5.728196620941162,
      "epoch": 2.009252539912917,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045964326829692925,
      "loss": 5.3983,
      "mean_token_accuracy": 0.17397216260433196,
      "num_tokens": 45965425.0,
      "step": 22150
    },
    {
      "entropy": 5.714250612258911,
      "epoch": 2.0097060957910013,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045962477630528064,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17838989943265915,
      "num_tokens": 45975637.0,
      "step": 22155
    },
    {
      "entropy": 5.671885585784912,
      "epoch": 2.0101596516690856,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045960628049554453,
      "loss": 5.4253,
      "mean_token_accuracy": 0.16598578095436095,
      "num_tokens": 45985268.0,
      "step": 22160
    },
    {
      "entropy": 5.656985282897949,
      "epoch": 2.01061320754717,
      "grad_norm": 2.25,
      "learning_rate": 0.00045958778086810354,
      "loss": 5.3001,
      "mean_token_accuracy": 0.17765935510396957,
      "num_tokens": 45995335.0,
      "step": 22165
    },
    {
      "entropy": 5.682910919189453,
      "epoch": 2.011066763425254,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004595692774233402,
      "loss": 5.4009,
      "mean_token_accuracy": 0.1661425933241844,
      "num_tokens": 46006037.0,
      "step": 22170
    },
    {
      "entropy": 5.74276762008667,
      "epoch": 2.011520319303338,
      "grad_norm": 2.375,
      "learning_rate": 0.00045955077016163713,
      "loss": 5.3399,
      "mean_token_accuracy": 0.17338268756866454,
      "num_tokens": 46017016.0,
      "step": 22175
    },
    {
      "entropy": 5.799929094314575,
      "epoch": 2.0119738751814222,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045953225908337717,
      "loss": 5.439,
      "mean_token_accuracy": 0.1677859902381897,
      "num_tokens": 46026797.0,
      "step": 22180
    },
    {
      "entropy": 5.681686639785767,
      "epoch": 2.0124274310595065,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045951374418894304,
      "loss": 5.3047,
      "mean_token_accuracy": 0.17850472629070283,
      "num_tokens": 46037799.0,
      "step": 22185
    },
    {
      "entropy": 5.684710073471069,
      "epoch": 2.0128809869375908,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045949522547871765,
      "loss": 5.2962,
      "mean_token_accuracy": 0.17344781756401062,
      "num_tokens": 46048617.0,
      "step": 22190
    },
    {
      "entropy": 5.725732851028442,
      "epoch": 2.013334542815675,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000459476702953084,
      "loss": 5.4251,
      "mean_token_accuracy": 0.16895399242639542,
      "num_tokens": 46060122.0,
      "step": 22195
    },
    {
      "entropy": 5.715791082382202,
      "epoch": 2.013788098693759,
      "grad_norm": 2.4375,
      "learning_rate": 0.000459458176612425,
      "loss": 5.3864,
      "mean_token_accuracy": 0.17435144484043122,
      "num_tokens": 46071488.0,
      "step": 22200
    },
    {
      "entropy": 5.747519063949585,
      "epoch": 2.014241654571843,
      "grad_norm": 1.875,
      "learning_rate": 0.00045943964645712407,
      "loss": 5.4416,
      "mean_token_accuracy": 0.16908272355794907,
      "num_tokens": 46081954.0,
      "step": 22205
    },
    {
      "entropy": 5.723187112808228,
      "epoch": 2.0146952104499274,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004594211124875642,
      "loss": 5.3509,
      "mean_token_accuracy": 0.1762673407793045,
      "num_tokens": 46092488.0,
      "step": 22210
    },
    {
      "entropy": 5.612676811218262,
      "epoch": 2.0151487663280117,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004594025747041287,
      "loss": 5.3309,
      "mean_token_accuracy": 0.18018595278263091,
      "num_tokens": 46102451.0,
      "step": 22215
    },
    {
      "entropy": 5.6796098232269285,
      "epoch": 2.015602322206096,
      "grad_norm": 2.15625,
      "learning_rate": 0.000459384033107201,
      "loss": 5.3921,
      "mean_token_accuracy": 0.17841228544712068,
      "num_tokens": 46113862.0,
      "step": 22220
    },
    {
      "entropy": 5.786471128463745,
      "epoch": 2.01605587808418,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004593654876971645,
      "loss": 5.4556,
      "mean_token_accuracy": 0.1673527866601944,
      "num_tokens": 46124517.0,
      "step": 22225
    },
    {
      "entropy": 5.654007816314698,
      "epoch": 2.016509433962264,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004593469384744027,
      "loss": 5.3833,
      "mean_token_accuracy": 0.17313019335269927,
      "num_tokens": 46134221.0,
      "step": 22230
    },
    {
      "entropy": 5.715832853317261,
      "epoch": 2.0169629898403483,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045932838543929924,
      "loss": 5.4162,
      "mean_token_accuracy": 0.1692165181040764,
      "num_tokens": 46144512.0,
      "step": 22235
    },
    {
      "entropy": 5.680310583114624,
      "epoch": 2.0174165457184325,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004593098285922378,
      "loss": 5.347,
      "mean_token_accuracy": 0.17413656115531922,
      "num_tokens": 46154373.0,
      "step": 22240
    },
    {
      "entropy": 5.711432266235351,
      "epoch": 2.017870101596517,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004592912679336021,
      "loss": 5.499,
      "mean_token_accuracy": 0.16913860142230988,
      "num_tokens": 46164834.0,
      "step": 22245
    },
    {
      "entropy": 5.732148599624634,
      "epoch": 2.018323657474601,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004592727034637761,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17703876048326492,
      "num_tokens": 46174815.0,
      "step": 22250
    },
    {
      "entropy": 5.69895339012146,
      "epoch": 2.018777213352685,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045925413518314363,
      "loss": 5.3618,
      "mean_token_accuracy": 0.16754695028066635,
      "num_tokens": 46185155.0,
      "step": 22255
    },
    {
      "entropy": 5.729181146621704,
      "epoch": 2.019230769230769,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004592355630920886,
      "loss": 5.3939,
      "mean_token_accuracy": 0.17464517503976823,
      "num_tokens": 46194537.0,
      "step": 22260
    },
    {
      "entropy": 5.766505527496338,
      "epoch": 2.0196843251088534,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045921698719099527,
      "loss": 5.4338,
      "mean_token_accuracy": 0.17167127430438994,
      "num_tokens": 46205494.0,
      "step": 22265
    },
    {
      "entropy": 5.6623166561126705,
      "epoch": 2.0201378809869377,
      "grad_norm": 2.828125,
      "learning_rate": 0.00045919840748024764,
      "loss": 5.3438,
      "mean_token_accuracy": 0.17703755646944047,
      "num_tokens": 46216399.0,
      "step": 22270
    },
    {
      "entropy": 5.587738370895385,
      "epoch": 2.020591436865022,
      "grad_norm": 2.640625,
      "learning_rate": 0.00045917982396023,
      "loss": 5.318,
      "mean_token_accuracy": 0.17534445226192474,
      "num_tokens": 46226515.0,
      "step": 22275
    },
    {
      "entropy": 5.7546289443969725,
      "epoch": 2.021044992743106,
      "grad_norm": 2.984375,
      "learning_rate": 0.00045916123663132676,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1691632404923439,
      "num_tokens": 46236688.0,
      "step": 22280
    },
    {
      "entropy": 5.713055419921875,
      "epoch": 2.02149854862119,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045914264549392215,
      "loss": 5.3555,
      "mean_token_accuracy": 0.18141342848539352,
      "num_tokens": 46247166.0,
      "step": 22285
    },
    {
      "entropy": 5.699989700317383,
      "epoch": 2.0219521044992743,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004591240505484007,
      "loss": 5.3537,
      "mean_token_accuracy": 0.17504015415906907,
      "num_tokens": 46258902.0,
      "step": 22290
    },
    {
      "entropy": 5.7037567615509035,
      "epoch": 2.0224056603773586,
      "grad_norm": 3.5,
      "learning_rate": 0.00045910545179514694,
      "loss": 5.4921,
      "mean_token_accuracy": 0.1709906354546547,
      "num_tokens": 46269819.0,
      "step": 22295
    },
    {
      "entropy": 5.8013303756713865,
      "epoch": 2.022859216255443,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004590868492345455,
      "loss": 5.4361,
      "mean_token_accuracy": 0.17790859043598176,
      "num_tokens": 46280715.0,
      "step": 22300
    },
    {
      "entropy": 5.719054126739502,
      "epoch": 2.0233127721335267,
      "grad_norm": 2.375,
      "learning_rate": 0.0004590682428669813,
      "loss": 5.4129,
      "mean_token_accuracy": 0.16961075365543365,
      "num_tokens": 46291079.0,
      "step": 22305
    },
    {
      "entropy": 5.699037885665893,
      "epoch": 2.023766328011611,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004590496326928387,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1762859344482422,
      "num_tokens": 46301362.0,
      "step": 22310
    },
    {
      "entropy": 5.747548341751099,
      "epoch": 2.024219883889695,
      "grad_norm": 3.140625,
      "learning_rate": 0.00045903101871250295,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17938854396343232,
      "num_tokens": 46311820.0,
      "step": 22315
    },
    {
      "entropy": 5.732650327682495,
      "epoch": 2.0246734397677795,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004590124009263587,
      "loss": 5.4567,
      "mean_token_accuracy": 0.17331631034612655,
      "num_tokens": 46321992.0,
      "step": 22320
    },
    {
      "entropy": 5.722734165191651,
      "epoch": 2.0251269956458637,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004589937793347912,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17292955070734023,
      "num_tokens": 46331537.0,
      "step": 22325
    },
    {
      "entropy": 5.742550897598266,
      "epoch": 2.0255805515239476,
      "grad_norm": 3.25,
      "learning_rate": 0.00045897515393818547,
      "loss": 5.3881,
      "mean_token_accuracy": 0.17360181659460067,
      "num_tokens": 46342214.0,
      "step": 22330
    },
    {
      "entropy": 5.641708755493164,
      "epoch": 2.026034107402032,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004589565247369266,
      "loss": 5.3574,
      "mean_token_accuracy": 0.17371378242969512,
      "num_tokens": 46352471.0,
      "step": 22335
    },
    {
      "entropy": 5.661126041412354,
      "epoch": 2.026487663280116,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004589378917314,
      "loss": 5.3916,
      "mean_token_accuracy": 0.1768624186515808,
      "num_tokens": 46363221.0,
      "step": 22340
    },
    {
      "entropy": 5.633500051498413,
      "epoch": 2.0269412191582004,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045891925492199087,
      "loss": 5.276,
      "mean_token_accuracy": 0.18581537008285523,
      "num_tokens": 46373454.0,
      "step": 22345
    },
    {
      "entropy": 5.791102409362793,
      "epoch": 2.0273947750362846,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004589006143090847,
      "loss": 5.5267,
      "mean_token_accuracy": 0.17159499377012252,
      "num_tokens": 46384403.0,
      "step": 22350
    },
    {
      "entropy": 5.677963352203369,
      "epoch": 2.0278483309143684,
      "grad_norm": 2.671875,
      "learning_rate": 0.00045888196989306697,
      "loss": 5.2514,
      "mean_token_accuracy": 0.18748920410871506,
      "num_tokens": 46393362.0,
      "step": 22355
    },
    {
      "entropy": 5.746590805053711,
      "epoch": 2.0283018867924527,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004588633216743232,
      "loss": 5.5036,
      "mean_token_accuracy": 0.17107126265764236,
      "num_tokens": 46405100.0,
      "step": 22360
    },
    {
      "entropy": 5.647452068328858,
      "epoch": 2.028755442670537,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045884466965323914,
      "loss": 5.3003,
      "mean_token_accuracy": 0.17596473097801207,
      "num_tokens": 46415058.0,
      "step": 22365
    },
    {
      "entropy": 5.657962131500244,
      "epoch": 2.0292089985486212,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045882601383020045,
      "loss": 5.3544,
      "mean_token_accuracy": 0.1761081784963608,
      "num_tokens": 46425863.0,
      "step": 22370
    },
    {
      "entropy": 5.7130539894104,
      "epoch": 2.0296625544267055,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004588073542055929,
      "loss": 5.3756,
      "mean_token_accuracy": 0.17288106232881545,
      "num_tokens": 46436172.0,
      "step": 22375
    },
    {
      "entropy": 5.761259651184082,
      "epoch": 2.0301161103047893,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045878869077980246,
      "loss": 5.3913,
      "mean_token_accuracy": 0.1710967853665352,
      "num_tokens": 46446717.0,
      "step": 22380
    },
    {
      "entropy": 5.739552354812622,
      "epoch": 2.0305696661828736,
      "grad_norm": 4.53125,
      "learning_rate": 0.00045877002355321504,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1689441293478012,
      "num_tokens": 46456944.0,
      "step": 22385
    },
    {
      "entropy": 5.728005123138428,
      "epoch": 2.031023222060958,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004587513525262168,
      "loss": 5.4495,
      "mean_token_accuracy": 0.16892146766185762,
      "num_tokens": 46467231.0,
      "step": 22390
    },
    {
      "entropy": 5.729822111129761,
      "epoch": 2.031476777939042,
      "grad_norm": 3.0625,
      "learning_rate": 0.00045873267769919364,
      "loss": 5.3145,
      "mean_token_accuracy": 0.16837610751390458,
      "num_tokens": 46477107.0,
      "step": 22395
    },
    {
      "entropy": 5.724275159835815,
      "epoch": 2.0319303338171264,
      "grad_norm": 2.5,
      "learning_rate": 0.0004587139990725319,
      "loss": 5.3724,
      "mean_token_accuracy": 0.17693231999874115,
      "num_tokens": 46487351.0,
      "step": 22400
    },
    {
      "entropy": 5.688820743560791,
      "epoch": 2.0323838896952107,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004586953166466179,
      "loss": 5.389,
      "mean_token_accuracy": 0.17397082448005677,
      "num_tokens": 46497685.0,
      "step": 22405
    },
    {
      "entropy": 5.669199800491333,
      "epoch": 2.0328374455732945,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004586766304218379,
      "loss": 5.3756,
      "mean_token_accuracy": 0.16709252446889877,
      "num_tokens": 46507866.0,
      "step": 22410
    },
    {
      "entropy": 5.619723320007324,
      "epoch": 2.0332910014513788,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045865794039857846,
      "loss": 5.2646,
      "mean_token_accuracy": 0.17970366328954696,
      "num_tokens": 46518797.0,
      "step": 22415
    },
    {
      "entropy": 5.6780038356781,
      "epoch": 2.033744557329463,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004586392465772259,
      "loss": 5.3369,
      "mean_token_accuracy": 0.17789751589298247,
      "num_tokens": 46528759.0,
      "step": 22420
    },
    {
      "entropy": 5.763638925552368,
      "epoch": 2.0341981132075473,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045862054895816703,
      "loss": 5.4853,
      "mean_token_accuracy": 0.16754979491233826,
      "num_tokens": 46539014.0,
      "step": 22425
    },
    {
      "entropy": 5.688062620162964,
      "epoch": 2.0346516690856316,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004586018475417884,
      "loss": 5.3417,
      "mean_token_accuracy": 0.17929524183273315,
      "num_tokens": 46548856.0,
      "step": 22430
    },
    {
      "entropy": 5.655853509902954,
      "epoch": 2.0351052249637154,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004585831423284766,
      "loss": 5.3016,
      "mean_token_accuracy": 0.17884412854909898,
      "num_tokens": 46559164.0,
      "step": 22435
    },
    {
      "entropy": 5.707827043533325,
      "epoch": 2.0355587808417996,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004585644333186188,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1717889279127121,
      "num_tokens": 46569793.0,
      "step": 22440
    },
    {
      "entropy": 5.703957414627075,
      "epoch": 2.036012336719884,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045854572051260166,
      "loss": 5.3852,
      "mean_token_accuracy": 0.16992079466581345,
      "num_tokens": 46580257.0,
      "step": 22445
    },
    {
      "entropy": 5.668719148635864,
      "epoch": 2.036465892597968,
      "grad_norm": 3.78125,
      "learning_rate": 0.00045852700391081234,
      "loss": 5.3778,
      "mean_token_accuracy": 0.17157640755176545,
      "num_tokens": 46590297.0,
      "step": 22450
    },
    {
      "entropy": 5.6964586734771725,
      "epoch": 2.0369194484760524,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004585082835136378,
      "loss": 5.3366,
      "mean_token_accuracy": 0.18089356869459153,
      "num_tokens": 46600647.0,
      "step": 22455
    },
    {
      "entropy": 5.694636249542237,
      "epoch": 2.0373730043541363,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004584895593214652,
      "loss": 5.3314,
      "mean_token_accuracy": 0.1741168439388275,
      "num_tokens": 46610478.0,
      "step": 22460
    },
    {
      "entropy": 5.640524530410767,
      "epoch": 2.0378265602322205,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045847083133468164,
      "loss": 5.4029,
      "mean_token_accuracy": 0.16721073389053345,
      "num_tokens": 46620159.0,
      "step": 22465
    },
    {
      "entropy": 5.713600015640258,
      "epoch": 2.038280116110305,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045845209955367464,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17716459631919862,
      "num_tokens": 46630555.0,
      "step": 22470
    },
    {
      "entropy": 5.743585252761841,
      "epoch": 2.038733671988389,
      "grad_norm": 3.953125,
      "learning_rate": 0.00045843336397883146,
      "loss": 5.4263,
      "mean_token_accuracy": 0.17193919718265532,
      "num_tokens": 46640188.0,
      "step": 22475
    },
    {
      "entropy": 5.697864246368408,
      "epoch": 2.0391872278664733,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004584146246105396,
      "loss": 5.3595,
      "mean_token_accuracy": 0.173255255818367,
      "num_tokens": 46650284.0,
      "step": 22480
    },
    {
      "entropy": 5.689089250564575,
      "epoch": 2.039640783744557,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045839588144918643,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1778315782546997,
      "num_tokens": 46661466.0,
      "step": 22485
    },
    {
      "entropy": 5.752255535125732,
      "epoch": 2.0400943396226414,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004583771344951598,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17068617939949035,
      "num_tokens": 46671683.0,
      "step": 22490
    },
    {
      "entropy": 5.699773550033569,
      "epoch": 2.0405478955007257,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045835838374884733,
      "loss": 5.396,
      "mean_token_accuracy": 0.17181251794099808,
      "num_tokens": 46681787.0,
      "step": 22495
    },
    {
      "entropy": 5.693995189666748,
      "epoch": 2.04100145137881,
      "grad_norm": 3.828125,
      "learning_rate": 0.00045833962921063665,
      "loss": 5.3751,
      "mean_token_accuracy": 0.1759103536605835,
      "num_tokens": 46692752.0,
      "step": 22500
    },
    {
      "entropy": 5.693674182891845,
      "epoch": 2.041455007256894,
      "grad_norm": 2.6875,
      "learning_rate": 0.00045832087088091574,
      "loss": 5.391,
      "mean_token_accuracy": 0.17193644344806672,
      "num_tokens": 46702980.0,
      "step": 22505
    },
    {
      "entropy": 5.65548267364502,
      "epoch": 2.041908563134978,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045830210876007257,
      "loss": 5.2848,
      "mean_token_accuracy": 0.1785035252571106,
      "num_tokens": 46712281.0,
      "step": 22510
    },
    {
      "entropy": 5.686748409271241,
      "epoch": 2.0423621190130623,
      "grad_norm": 2.234375,
      "learning_rate": 0.000458283342848495,
      "loss": 5.3357,
      "mean_token_accuracy": 0.17701264172792436,
      "num_tokens": 46722097.0,
      "step": 22515
    },
    {
      "entropy": 5.6321690559387205,
      "epoch": 2.0428156748911466,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004582645731465713,
      "loss": 5.2799,
      "mean_token_accuracy": 0.17957015186548234,
      "num_tokens": 46733056.0,
      "step": 22520
    },
    {
      "entropy": 5.618005180358887,
      "epoch": 2.043269230769231,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045824579965468944,
      "loss": 5.3159,
      "mean_token_accuracy": 0.17507190853357316,
      "num_tokens": 46742608.0,
      "step": 22525
    },
    {
      "entropy": 5.669701433181762,
      "epoch": 2.043722786647315,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045822702237323777,
      "loss": 5.2884,
      "mean_token_accuracy": 0.1788678362965584,
      "num_tokens": 46751762.0,
      "step": 22530
    },
    {
      "entropy": 5.683850860595703,
      "epoch": 2.044176342525399,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045820824130260457,
      "loss": 5.2984,
      "mean_token_accuracy": 0.1852424368262291,
      "num_tokens": 46762252.0,
      "step": 22535
    },
    {
      "entropy": 5.6851438045501705,
      "epoch": 2.044629898403483,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004581894564431783,
      "loss": 5.3676,
      "mean_token_accuracy": 0.17212578654289246,
      "num_tokens": 46772172.0,
      "step": 22540
    },
    {
      "entropy": 5.689287567138672,
      "epoch": 2.0450834542815675,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004581706677953473,
      "loss": 5.3904,
      "mean_token_accuracy": 0.17346883416175843,
      "num_tokens": 46782894.0,
      "step": 22545
    },
    {
      "entropy": 5.701255130767822,
      "epoch": 2.0455370101596517,
      "grad_norm": 3.125,
      "learning_rate": 0.00045815187535950034,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17156564593315124,
      "num_tokens": 46793097.0,
      "step": 22550
    },
    {
      "entropy": 5.787491798400879,
      "epoch": 2.045990566037736,
      "grad_norm": 3.234375,
      "learning_rate": 0.00045813307913602586,
      "loss": 5.4795,
      "mean_token_accuracy": 0.16375991702079773,
      "num_tokens": 46804196.0,
      "step": 22555
    },
    {
      "entropy": 5.7156726837158205,
      "epoch": 2.04644412191582,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004581142791253126,
      "loss": 5.3609,
      "mean_token_accuracy": 0.17552577555179597,
      "num_tokens": 46813985.0,
      "step": 22560
    },
    {
      "entropy": 5.643201684951782,
      "epoch": 2.046897677793904,
      "grad_norm": 2.640625,
      "learning_rate": 0.00045809547532774943,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17581314742565154,
      "num_tokens": 46824125.0,
      "step": 22565
    },
    {
      "entropy": 5.637062358856201,
      "epoch": 2.0473512336719883,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045807666774372514,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17941786646842955,
      "num_tokens": 46834982.0,
      "step": 22570
    },
    {
      "entropy": 5.638332033157349,
      "epoch": 2.0478047895500726,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045805785637362874,
      "loss": 5.2838,
      "mean_token_accuracy": 0.18527984172105788,
      "num_tokens": 46844669.0,
      "step": 22575
    },
    {
      "entropy": 5.679537010192871,
      "epoch": 2.048258345428157,
      "grad_norm": 2.625,
      "learning_rate": 0.0004580390412178492,
      "loss": 5.3852,
      "mean_token_accuracy": 0.18212694376707078,
      "num_tokens": 46855108.0,
      "step": 22580
    },
    {
      "entropy": 5.761588525772095,
      "epoch": 2.048711901306241,
      "grad_norm": 2.359375,
      "learning_rate": 0.00045802022227677566,
      "loss": 5.3174,
      "mean_token_accuracy": 0.17632479816675187,
      "num_tokens": 46864878.0,
      "step": 22585
    },
    {
      "entropy": 5.712237310409546,
      "epoch": 2.049165457184325,
      "grad_norm": 4.53125,
      "learning_rate": 0.00045800139955079727,
      "loss": 5.3776,
      "mean_token_accuracy": 0.17299358546733856,
      "num_tokens": 46874916.0,
      "step": 22590
    },
    {
      "entropy": 5.683449745178223,
      "epoch": 2.0496190130624092,
      "grad_norm": 2.75,
      "learning_rate": 0.0004579825730403033,
      "loss": 5.4306,
      "mean_token_accuracy": 0.1705819308757782,
      "num_tokens": 46885167.0,
      "step": 22595
    },
    {
      "entropy": 5.713363790512085,
      "epoch": 2.0500725689404935,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004579637427456831,
      "loss": 5.3738,
      "mean_token_accuracy": 0.1712704196572304,
      "num_tokens": 46895526.0,
      "step": 22600
    },
    {
      "entropy": 5.679062414169311,
      "epoch": 2.0505261248185778,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045794490866732607,
      "loss": 5.3622,
      "mean_token_accuracy": 0.17284103184938432,
      "num_tokens": 46905426.0,
      "step": 22605
    },
    {
      "entropy": 5.72233099937439,
      "epoch": 2.050979680696662,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004579260708056217,
      "loss": 5.3656,
      "mean_token_accuracy": 0.17639652043581008,
      "num_tokens": 46915728.0,
      "step": 22610
    },
    {
      "entropy": 5.6624696254730225,
      "epoch": 2.051433236574746,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004579072291609596,
      "loss": 5.3125,
      "mean_token_accuracy": 0.1760306775569916,
      "num_tokens": 46925972.0,
      "step": 22615
    },
    {
      "entropy": 5.683280181884766,
      "epoch": 2.05188679245283,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045788838373372935,
      "loss": 5.3783,
      "mean_token_accuracy": 0.18243319541215897,
      "num_tokens": 46937108.0,
      "step": 22620
    },
    {
      "entropy": 5.667391920089722,
      "epoch": 2.0523403483309144,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045786953452432076,
      "loss": 5.3334,
      "mean_token_accuracy": 0.17303056865930558,
      "num_tokens": 46947684.0,
      "step": 22625
    },
    {
      "entropy": 5.7347487449646,
      "epoch": 2.0527939042089987,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004578506815331236,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16640390008687972,
      "num_tokens": 46958089.0,
      "step": 22630
    },
    {
      "entropy": 5.676932764053345,
      "epoch": 2.053247460087083,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004578318247605277,
      "loss": 5.3252,
      "mean_token_accuracy": 0.18075264543294906,
      "num_tokens": 46968618.0,
      "step": 22635
    },
    {
      "entropy": 5.6997339725494385,
      "epoch": 2.0537010159651667,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004578129642069231,
      "loss": 5.3587,
      "mean_token_accuracy": 0.17634198665618897,
      "num_tokens": 46978791.0,
      "step": 22640
    },
    {
      "entropy": 5.721356534957886,
      "epoch": 2.054154571843251,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004577940998726998,
      "loss": 5.29,
      "mean_token_accuracy": 0.1797158181667328,
      "num_tokens": 46988706.0,
      "step": 22645
    },
    {
      "entropy": 5.7495826244354244,
      "epoch": 2.0546081277213353,
      "grad_norm": 2.890625,
      "learning_rate": 0.00045777523175824796,
      "loss": 5.5014,
      "mean_token_accuracy": 0.1571715459227562,
      "num_tokens": 47000140.0,
      "step": 22650
    },
    {
      "entropy": 5.701590967178345,
      "epoch": 2.0550616835994195,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004577563598639577,
      "loss": 5.3284,
      "mean_token_accuracy": 0.18053314387798308,
      "num_tokens": 47009773.0,
      "step": 22655
    },
    {
      "entropy": 5.765002298355102,
      "epoch": 2.055515239477504,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045773748419021944,
      "loss": 5.3968,
      "mean_token_accuracy": 0.16884640008211135,
      "num_tokens": 47020326.0,
      "step": 22660
    },
    {
      "entropy": 5.649790620803833,
      "epoch": 2.0559687953555876,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00045771860473742347,
      "loss": 5.3228,
      "mean_token_accuracy": 0.17235369831323624,
      "num_tokens": 47030410.0,
      "step": 22665
    },
    {
      "entropy": 5.684008407592773,
      "epoch": 2.056422351233672,
      "grad_norm": 1.75,
      "learning_rate": 0.00045769972150596006,
      "loss": 5.3817,
      "mean_token_accuracy": 0.1719810903072357,
      "num_tokens": 47040546.0,
      "step": 22670
    },
    {
      "entropy": 5.729417705535889,
      "epoch": 2.056875907111756,
      "grad_norm": 3.125,
      "learning_rate": 0.00045768083449621987,
      "loss": 5.3438,
      "mean_token_accuracy": 0.17992307096719742,
      "num_tokens": 47050210.0,
      "step": 22675
    },
    {
      "entropy": 5.774865818023682,
      "epoch": 2.0573294629898404,
      "grad_norm": 2.65625,
      "learning_rate": 0.00045766194370859347,
      "loss": 5.4882,
      "mean_token_accuracy": 0.16202962324023246,
      "num_tokens": 47060641.0,
      "step": 22680
    },
    {
      "entropy": 5.713528490066528,
      "epoch": 2.0577830188679247,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004576430491434716,
      "loss": 5.4215,
      "mean_token_accuracy": 0.17433919459581376,
      "num_tokens": 47070737.0,
      "step": 22685
    },
    {
      "entropy": 5.78364930152893,
      "epoch": 2.0582365747460085,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045762415080124487,
      "loss": 5.3823,
      "mean_token_accuracy": 0.17339835911989213,
      "num_tokens": 47080884.0,
      "step": 22690
    },
    {
      "entropy": 5.688161706924438,
      "epoch": 2.058690130624093,
      "grad_norm": 3.4375,
      "learning_rate": 0.00045760524868230417,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17452422529459,
      "num_tokens": 47090865.0,
      "step": 22695
    },
    {
      "entropy": 5.679276752471924,
      "epoch": 2.059143686502177,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004575863427870404,
      "loss": 5.3659,
      "mean_token_accuracy": 0.17780518978834153,
      "num_tokens": 47100570.0,
      "step": 22700
    },
    {
      "entropy": 5.707064771652222,
      "epoch": 2.0595972423802613,
      "grad_norm": 4.875,
      "learning_rate": 0.0004575674331158446,
      "loss": 5.367,
      "mean_token_accuracy": 0.18101715594530104,
      "num_tokens": 47111199.0,
      "step": 22705
    },
    {
      "entropy": 5.689499425888061,
      "epoch": 2.0600507982583456,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004575485196691076,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17675517201423646,
      "num_tokens": 47121191.0,
      "step": 22710
    },
    {
      "entropy": 5.724070835113525,
      "epoch": 2.0605043541364294,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004575296024472209,
      "loss": 5.4519,
      "mean_token_accuracy": 0.171602700650692,
      "num_tokens": 47131608.0,
      "step": 22715
    },
    {
      "entropy": 5.746758604049683,
      "epoch": 2.0609579100145137,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004575106814505753,
      "loss": 5.3781,
      "mean_token_accuracy": 0.17099542915821075,
      "num_tokens": 47141492.0,
      "step": 22720
    },
    {
      "entropy": 5.6902796745300295,
      "epoch": 2.061411465892598,
      "grad_norm": 2.75,
      "learning_rate": 0.0004574917566795624,
      "loss": 5.2721,
      "mean_token_accuracy": 0.1804874286055565,
      "num_tokens": 47152576.0,
      "step": 22725
    },
    {
      "entropy": 5.70591893196106,
      "epoch": 2.061865021770682,
      "grad_norm": 2.734375,
      "learning_rate": 0.00045747282813457345,
      "loss": 5.3878,
      "mean_token_accuracy": 0.172776560485363,
      "num_tokens": 47163863.0,
      "step": 22730
    },
    {
      "entropy": 5.690427684783936,
      "epoch": 2.0623185776487665,
      "grad_norm": 2.0,
      "learning_rate": 0.000457453895816,
      "loss": 5.3374,
      "mean_token_accuracy": 0.1703617051243782,
      "num_tokens": 47174140.0,
      "step": 22735
    },
    {
      "entropy": 5.747409725189209,
      "epoch": 2.0627721335268503,
      "grad_norm": 3.46875,
      "learning_rate": 0.00045743495972423346,
      "loss": 5.3719,
      "mean_token_accuracy": 0.17258910089731216,
      "num_tokens": 47184708.0,
      "step": 22740
    },
    {
      "entropy": 5.704762506484985,
      "epoch": 2.0632256894049346,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004574160198596653,
      "loss": 5.3809,
      "mean_token_accuracy": 0.16964446157217025,
      "num_tokens": 47196286.0,
      "step": 22745
    },
    {
      "entropy": 5.5879981994628904,
      "epoch": 2.063679245283019,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045739707622268757,
      "loss": 5.3329,
      "mean_token_accuracy": 0.18036781549453734,
      "num_tokens": 47206774.0,
      "step": 22750
    },
    {
      "entropy": 5.633327150344849,
      "epoch": 2.064132801161103,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004573781288136917,
      "loss": 5.2526,
      "mean_token_accuracy": 0.17920863032341003,
      "num_tokens": 47217009.0,
      "step": 22755
    },
    {
      "entropy": 5.670976781845093,
      "epoch": 2.0645863570391874,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045735917763306967,
      "loss": 5.3732,
      "mean_token_accuracy": 0.1725206822156906,
      "num_tokens": 47226718.0,
      "step": 22760
    },
    {
      "entropy": 5.599688577651977,
      "epoch": 2.065039912917271,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004573402226812134,
      "loss": 5.2525,
      "mean_token_accuracy": 0.180233696103096,
      "num_tokens": 47237666.0,
      "step": 22765
    },
    {
      "entropy": 5.63920030593872,
      "epoch": 2.0654934687953554,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045732126395851483,
      "loss": 5.4244,
      "mean_token_accuracy": 0.1697695627808571,
      "num_tokens": 47249230.0,
      "step": 22770
    },
    {
      "entropy": 5.717934036254883,
      "epoch": 2.0659470246734397,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004573023014653661,
      "loss": 5.317,
      "mean_token_accuracy": 0.18244122117757797,
      "num_tokens": 47259232.0,
      "step": 22775
    },
    {
      "entropy": 5.7179313659667965,
      "epoch": 2.066400580551524,
      "grad_norm": 1.921875,
      "learning_rate": 0.00045728333520215933,
      "loss": 5.3601,
      "mean_token_accuracy": 0.17527539730072023,
      "num_tokens": 47269006.0,
      "step": 22780
    },
    {
      "entropy": 5.714682912826538,
      "epoch": 2.0668541364296082,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004572643651692867,
      "loss": 5.4816,
      "mean_token_accuracy": 0.16443315893411636,
      "num_tokens": 47280048.0,
      "step": 22785
    },
    {
      "entropy": 5.641515445709229,
      "epoch": 2.0673076923076925,
      "grad_norm": 3.125,
      "learning_rate": 0.00045724539136714063,
      "loss": 5.3095,
      "mean_token_accuracy": 0.175929556787014,
      "num_tokens": 47290019.0,
      "step": 22790
    },
    {
      "entropy": 5.713611745834351,
      "epoch": 2.0677612481857763,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004572264137961133,
      "loss": 5.3298,
      "mean_token_accuracy": 0.18017607033252717,
      "num_tokens": 47300573.0,
      "step": 22795
    },
    {
      "entropy": 5.7124754905700685,
      "epoch": 2.0682148040638606,
      "grad_norm": 2.34375,
      "learning_rate": 0.00045720743245659734,
      "loss": 5.3804,
      "mean_token_accuracy": 0.16796911656856536,
      "num_tokens": 47311795.0,
      "step": 22800
    },
    {
      "entropy": 5.6065596580505375,
      "epoch": 2.068668359941945,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004571884473489853,
      "loss": 5.2503,
      "mean_token_accuracy": 0.18653201013803483,
      "num_tokens": 47321768.0,
      "step": 22805
    },
    {
      "entropy": 5.713085746765136,
      "epoch": 2.069121915820029,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004571694584736696,
      "loss": 5.4372,
      "mean_token_accuracy": 0.1743260145187378,
      "num_tokens": 47332662.0,
      "step": 22810
    },
    {
      "entropy": 5.7073300838470455,
      "epoch": 2.0695754716981134,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004571504658310432,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17126732915639878,
      "num_tokens": 47344355.0,
      "step": 22815
    },
    {
      "entropy": 5.668503284454346,
      "epoch": 2.0700290275761972,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004571314694214987,
      "loss": 5.297,
      "mean_token_accuracy": 0.17856367975473403,
      "num_tokens": 47354228.0,
      "step": 22820
    },
    {
      "entropy": 5.706161260604858,
      "epoch": 2.0704825834542815,
      "grad_norm": 6.25,
      "learning_rate": 0.0004571124692454291,
      "loss": 5.3703,
      "mean_token_accuracy": 0.1783539965748787,
      "num_tokens": 47364477.0,
      "step": 22825
    },
    {
      "entropy": 5.767140531539917,
      "epoch": 2.0709361393323658,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045709346530322715,
      "loss": 5.4134,
      "mean_token_accuracy": 0.17540447264909745,
      "num_tokens": 47375745.0,
      "step": 22830
    },
    {
      "entropy": 5.711370420455933,
      "epoch": 2.07138969521045,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004570744575952859,
      "loss": 5.4372,
      "mean_token_accuracy": 0.17852006405591964,
      "num_tokens": 47385341.0,
      "step": 22835
    },
    {
      "entropy": 5.657266569137573,
      "epoch": 2.0718432510885343,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004570554461219985,
      "loss": 5.3522,
      "mean_token_accuracy": 0.1783701255917549,
      "num_tokens": 47396266.0,
      "step": 22840
    },
    {
      "entropy": 5.708980655670166,
      "epoch": 2.072296806966618,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045703643088375806,
      "loss": 5.3499,
      "mean_token_accuracy": 0.171886083483696,
      "num_tokens": 47406772.0,
      "step": 22845
    },
    {
      "entropy": 5.748170280456543,
      "epoch": 2.0727503628447024,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045701741188095793,
      "loss": 5.4405,
      "mean_token_accuracy": 0.17067295014858247,
      "num_tokens": 47416833.0,
      "step": 22850
    },
    {
      "entropy": 5.680977201461792,
      "epoch": 2.0732039187227866,
      "grad_norm": 4.125,
      "learning_rate": 0.0004569983891139912,
      "loss": 5.3499,
      "mean_token_accuracy": 0.17959507554769516,
      "num_tokens": 47426954.0,
      "step": 22855
    },
    {
      "entropy": 5.585741281509399,
      "epoch": 2.073657474600871,
      "grad_norm": 2.875,
      "learning_rate": 0.00045697936258325155,
      "loss": 5.2959,
      "mean_token_accuracy": 0.18235497176647186,
      "num_tokens": 47437520.0,
      "step": 22860
    },
    {
      "entropy": 5.678774118423462,
      "epoch": 2.074111030478955,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045696033228913226,
      "loss": 5.3195,
      "mean_token_accuracy": 0.17469614446163179,
      "num_tokens": 47447977.0,
      "step": 22865
    },
    {
      "entropy": 5.774969387054443,
      "epoch": 2.074564586357039,
      "grad_norm": 4.34375,
      "learning_rate": 0.00045694129823202684,
      "loss": 5.3966,
      "mean_token_accuracy": 0.1729213312268257,
      "num_tokens": 47458458.0,
      "step": 22870
    },
    {
      "entropy": 5.698882913589477,
      "epoch": 2.0750181422351233,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004569222604123291,
      "loss": 5.3837,
      "mean_token_accuracy": 0.17182944416999818,
      "num_tokens": 47469048.0,
      "step": 22875
    },
    {
      "entropy": 5.713098478317261,
      "epoch": 2.0754716981132075,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004569032188304326,
      "loss": 5.441,
      "mean_token_accuracy": 0.16607949882745743,
      "num_tokens": 47480353.0,
      "step": 22880
    },
    {
      "entropy": 5.65142936706543,
      "epoch": 2.075925253991292,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004568841734867313,
      "loss": 5.2536,
      "mean_token_accuracy": 0.17905064672231674,
      "num_tokens": 47489951.0,
      "step": 22885
    },
    {
      "entropy": 5.651728582382202,
      "epoch": 2.076378809869376,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004568651243816188,
      "loss": 5.3013,
      "mean_token_accuracy": 0.18016261607408524,
      "num_tokens": 47499015.0,
      "step": 22890
    },
    {
      "entropy": 5.747555541992187,
      "epoch": 2.07683236574746,
      "grad_norm": 3.6875,
      "learning_rate": 0.00045684607151548923,
      "loss": 5.354,
      "mean_token_accuracy": 0.18095378279685975,
      "num_tokens": 47508694.0,
      "step": 22895
    },
    {
      "entropy": 5.860283422470093,
      "epoch": 2.077285921625544,
      "grad_norm": 8.1875,
      "learning_rate": 0.00045682701488873655,
      "loss": 5.5261,
      "mean_token_accuracy": 0.1628246396780014,
      "num_tokens": 47521336.0,
      "step": 22900
    },
    {
      "entropy": 5.792023992538452,
      "epoch": 2.0777394775036284,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004568079545017549,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16394307613372802,
      "num_tokens": 47531857.0,
      "step": 22905
    },
    {
      "entropy": 5.703441905975342,
      "epoch": 2.0781930333817127,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004567888903549385,
      "loss": 5.3754,
      "mean_token_accuracy": 0.16536602824926377,
      "num_tokens": 47542224.0,
      "step": 22910
    },
    {
      "entropy": 5.687231731414795,
      "epoch": 2.078646589259797,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004567698224486813,
      "loss": 5.4297,
      "mean_token_accuracy": 0.1745286375284195,
      "num_tokens": 47552370.0,
      "step": 22915
    },
    {
      "entropy": 5.803608703613281,
      "epoch": 2.0791001451378808,
      "grad_norm": 2.796875,
      "learning_rate": 0.00045675075078337803,
      "loss": 5.3621,
      "mean_token_accuracy": 0.17919624000787734,
      "num_tokens": 47561997.0,
      "step": 22920
    },
    {
      "entropy": 5.8260640621185305,
      "epoch": 2.079553701015965,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004567316753594229,
      "loss": 5.3622,
      "mean_token_accuracy": 0.1788135811686516,
      "num_tokens": 47571968.0,
      "step": 22925
    },
    {
      "entropy": 5.773275756835938,
      "epoch": 2.0800072568940493,
      "grad_norm": 3.421875,
      "learning_rate": 0.00045671259617721037,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17015140652656555,
      "num_tokens": 47583459.0,
      "step": 22930
    },
    {
      "entropy": 5.63952865600586,
      "epoch": 2.0804608127721336,
      "grad_norm": 2.75,
      "learning_rate": 0.000456693513237135,
      "loss": 5.3067,
      "mean_token_accuracy": 0.17538341134786606,
      "num_tokens": 47594026.0,
      "step": 22935
    },
    {
      "entropy": 5.73462233543396,
      "epoch": 2.080914368650218,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004566744265395916,
      "loss": 5.4016,
      "mean_token_accuracy": 0.1710807204246521,
      "num_tokens": 47605246.0,
      "step": 22940
    },
    {
      "entropy": 5.748352336883545,
      "epoch": 2.081367924528302,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045665533608497467,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16837071180343627,
      "num_tokens": 47616702.0,
      "step": 22945
    },
    {
      "entropy": 5.81542387008667,
      "epoch": 2.081821480406386,
      "grad_norm": 2.625,
      "learning_rate": 0.00045663624187367914,
      "loss": 5.4497,
      "mean_token_accuracy": 0.16723241060972213,
      "num_tokens": 47627616.0,
      "step": 22950
    },
    {
      "entropy": 5.716643857955932,
      "epoch": 2.08227503628447,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004566171439060998,
      "loss": 5.3317,
      "mean_token_accuracy": 0.18068880140781401,
      "num_tokens": 47637287.0,
      "step": 22955
    },
    {
      "entropy": 5.61277585029602,
      "epoch": 2.0827285921625545,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004565980421826317,
      "loss": 5.3367,
      "mean_token_accuracy": 0.17799465507268905,
      "num_tokens": 47647474.0,
      "step": 22960
    },
    {
      "entropy": 5.739857339859009,
      "epoch": 2.0831821480406387,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004565789367036697,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17650084793567658,
      "num_tokens": 47657036.0,
      "step": 22965
    },
    {
      "entropy": 5.6945929527282715,
      "epoch": 2.083635703918723,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004565598274696091,
      "loss": 5.2947,
      "mean_token_accuracy": 0.17640351504087448,
      "num_tokens": 47668026.0,
      "step": 22970
    },
    {
      "entropy": 5.822618865966797,
      "epoch": 2.084089259796807,
      "grad_norm": 3.15625,
      "learning_rate": 0.00045654071448084506,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1685180887579918,
      "num_tokens": 47679508.0,
      "step": 22975
    },
    {
      "entropy": 5.735685539245606,
      "epoch": 2.084542815674891,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045652159773777273,
      "loss": 5.3624,
      "mean_token_accuracy": 0.175012069940567,
      "num_tokens": 47688474.0,
      "step": 22980
    },
    {
      "entropy": 5.767128801345825,
      "epoch": 2.0849963715529753,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004565024772407875,
      "loss": 5.5595,
      "mean_token_accuracy": 0.15873113870620728,
      "num_tokens": 47698675.0,
      "step": 22985
    },
    {
      "entropy": 5.7133691787719725,
      "epoch": 2.0854499274310596,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004564833529902848,
      "loss": 5.3608,
      "mean_token_accuracy": 0.18450791090726854,
      "num_tokens": 47709214.0,
      "step": 22990
    },
    {
      "entropy": 5.763336420059204,
      "epoch": 2.085903483309144,
      "grad_norm": 9.1875,
      "learning_rate": 0.00045646422498666,
      "loss": 5.4625,
      "mean_token_accuracy": 0.17127978056669235,
      "num_tokens": 47721000.0,
      "step": 22995
    },
    {
      "entropy": 5.661542272567749,
      "epoch": 2.0863570391872277,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045644509323030893,
      "loss": 5.3504,
      "mean_token_accuracy": 0.17520327866077423,
      "num_tokens": 47731218.0,
      "step": 23000
    },
    {
      "entropy": 5.680297613143921,
      "epoch": 2.086810595065312,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045642595772162704,
      "loss": 5.35,
      "mean_token_accuracy": 0.17445975691080093,
      "num_tokens": 47742339.0,
      "step": 23005
    },
    {
      "entropy": 5.723939418792725,
      "epoch": 2.0872641509433962,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045640681846101014,
      "loss": 5.4474,
      "mean_token_accuracy": 0.17067268639802932,
      "num_tokens": 47752858.0,
      "step": 23010
    },
    {
      "entropy": 5.661289691925049,
      "epoch": 2.0877177068214805,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004563876754488539,
      "loss": 5.2979,
      "mean_token_accuracy": 0.18352532833814622,
      "num_tokens": 47763303.0,
      "step": 23015
    },
    {
      "entropy": 5.696356868743896,
      "epoch": 2.0881712626995648,
      "grad_norm": 2.625,
      "learning_rate": 0.0004563685286855544,
      "loss": 5.3391,
      "mean_token_accuracy": 0.17365500777959825,
      "num_tokens": 47773212.0,
      "step": 23020
    },
    {
      "entropy": 5.75771131515503,
      "epoch": 2.0886248185776486,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004563493781715075,
      "loss": 5.4747,
      "mean_token_accuracy": 0.1704810544848442,
      "num_tokens": 47783376.0,
      "step": 23025
    },
    {
      "entropy": 5.675118398666382,
      "epoch": 2.089078374455733,
      "grad_norm": 2.75,
      "learning_rate": 0.0004563302239071093,
      "loss": 5.3665,
      "mean_token_accuracy": 0.1775444507598877,
      "num_tokens": 47794859.0,
      "step": 23030
    },
    {
      "entropy": 5.72062840461731,
      "epoch": 2.089531930333817,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045631106589275576,
      "loss": 5.3838,
      "mean_token_accuracy": 0.17606519907712936,
      "num_tokens": 47804721.0,
      "step": 23035
    },
    {
      "entropy": 5.711503791809082,
      "epoch": 2.0899854862119014,
      "grad_norm": 2.65625,
      "learning_rate": 0.00045629190412884324,
      "loss": 5.2771,
      "mean_token_accuracy": 0.18160481750965118,
      "num_tokens": 47814701.0,
      "step": 23040
    },
    {
      "entropy": 5.664831829071045,
      "epoch": 2.0904390420899857,
      "grad_norm": 3.03125,
      "learning_rate": 0.00045627273861576786,
      "loss": 5.2946,
      "mean_token_accuracy": 0.18355926871299744,
      "num_tokens": 47823941.0,
      "step": 23045
    },
    {
      "entropy": 5.713161373138428,
      "epoch": 2.0908925979680695,
      "grad_norm": 2.34375,
      "learning_rate": 0.00045625356935392614,
      "loss": 5.4114,
      "mean_token_accuracy": 0.16732460856437684,
      "num_tokens": 47835793.0,
      "step": 23050
    },
    {
      "entropy": 5.7970785140991214,
      "epoch": 2.0913461538461537,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004562343963437144,
      "loss": 5.4358,
      "mean_token_accuracy": 0.17130852043628692,
      "num_tokens": 47845677.0,
      "step": 23055
    },
    {
      "entropy": 5.760763883590698,
      "epoch": 2.091799709724238,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004562152195855291,
      "loss": 5.3423,
      "mean_token_accuracy": 0.1822948470711708,
      "num_tokens": 47855611.0,
      "step": 23060
    },
    {
      "entropy": 5.694050216674805,
      "epoch": 2.0922532656023223,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004561960390797669,
      "loss": 5.3578,
      "mean_token_accuracy": 0.17453897446393968,
      "num_tokens": 47865115.0,
      "step": 23065
    },
    {
      "entropy": 5.804679536819458,
      "epoch": 2.0927068214804065,
      "grad_norm": 3.25,
      "learning_rate": 0.00045617685482682454,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16932236552238464,
      "num_tokens": 47876690.0,
      "step": 23070
    },
    {
      "entropy": 5.687907028198242,
      "epoch": 2.0931603773584904,
      "grad_norm": 3.0,
      "learning_rate": 0.0004561576668270986,
      "loss": 5.3712,
      "mean_token_accuracy": 0.17692483961582184,
      "num_tokens": 47887709.0,
      "step": 23075
    },
    {
      "entropy": 5.676269483566284,
      "epoch": 2.0936139332365746,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045613847508098587,
      "loss": 5.3735,
      "mean_token_accuracy": 0.1761247083544731,
      "num_tokens": 47898209.0,
      "step": 23080
    },
    {
      "entropy": 5.672044944763184,
      "epoch": 2.094067489114659,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045611927958888335,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1811466857790947,
      "num_tokens": 47907494.0,
      "step": 23085
    },
    {
      "entropy": 5.6509418964385985,
      "epoch": 2.094521044992743,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045610008035118797,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17840507626533508,
      "num_tokens": 47916923.0,
      "step": 23090
    },
    {
      "entropy": 5.616380405426026,
      "epoch": 2.0949746008708274,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045608087736829687,
      "loss": 5.3164,
      "mean_token_accuracy": 0.18090560883283616,
      "num_tokens": 47927043.0,
      "step": 23095
    },
    {
      "entropy": 5.762189435958862,
      "epoch": 2.0954281567489113,
      "grad_norm": 2.578125,
      "learning_rate": 0.000456061670640607,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17442049235105514,
      "num_tokens": 47937296.0,
      "step": 23100
    },
    {
      "entropy": 5.77219557762146,
      "epoch": 2.0958817126269955,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004560424601685157,
      "loss": 5.4119,
      "mean_token_accuracy": 0.17585094571113585,
      "num_tokens": 47948268.0,
      "step": 23105
    },
    {
      "entropy": 5.722855138778686,
      "epoch": 2.09633526850508,
      "grad_norm": 2.0,
      "learning_rate": 0.0004560232459524201,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1749066650867462,
      "num_tokens": 47958375.0,
      "step": 23110
    },
    {
      "entropy": 5.739527225494385,
      "epoch": 2.096788824383164,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045600402799271766,
      "loss": 5.4051,
      "mean_token_accuracy": 0.18133500069379807,
      "num_tokens": 47968469.0,
      "step": 23115
    },
    {
      "entropy": 5.6940210342407225,
      "epoch": 2.0972423802612483,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004559848062898058,
      "loss": 5.318,
      "mean_token_accuracy": 0.18000999242067336,
      "num_tokens": 47979630.0,
      "step": 23120
    },
    {
      "entropy": 5.710601043701172,
      "epoch": 2.097695936139332,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004559655808440821,
      "loss": 5.3504,
      "mean_token_accuracy": 0.17629560977220535,
      "num_tokens": 47989828.0,
      "step": 23125
    },
    {
      "entropy": 5.729275274276733,
      "epoch": 2.0981494920174164,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045594635165594404,
      "loss": 5.4704,
      "mean_token_accuracy": 0.16594031900167466,
      "num_tokens": 47999219.0,
      "step": 23130
    },
    {
      "entropy": 5.742139387130737,
      "epoch": 2.0986030478955007,
      "grad_norm": 2.203125,
      "learning_rate": 0.00045592711872578925,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1831235557794571,
      "num_tokens": 48009244.0,
      "step": 23135
    },
    {
      "entropy": 5.70686149597168,
      "epoch": 2.099056603773585,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045590788205401557,
      "loss": 5.3124,
      "mean_token_accuracy": 0.17627809196710587,
      "num_tokens": 48019066.0,
      "step": 23140
    },
    {
      "entropy": 5.604750156402588,
      "epoch": 2.099510159651669,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045588864164102086,
      "loss": 5.318,
      "mean_token_accuracy": 0.18179002404212952,
      "num_tokens": 48028704.0,
      "step": 23145
    },
    {
      "entropy": 5.676843595504761,
      "epoch": 2.0999637155297535,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045586939748720284,
      "loss": 5.3352,
      "mean_token_accuracy": 0.18212227821350097,
      "num_tokens": 48039210.0,
      "step": 23150
    },
    {
      "entropy": 5.707198810577393,
      "epoch": 2.1004172714078373,
      "grad_norm": 2.953125,
      "learning_rate": 0.00045585014959295965,
      "loss": 5.4576,
      "mean_token_accuracy": 0.1690465912222862,
      "num_tokens": 48048446.0,
      "step": 23155
    },
    {
      "entropy": 5.728558874130249,
      "epoch": 2.1008708272859216,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045583089795868934,
      "loss": 5.374,
      "mean_token_accuracy": 0.17779878228902818,
      "num_tokens": 48058880.0,
      "step": 23160
    },
    {
      "entropy": 5.7341564178466795,
      "epoch": 2.101324383164006,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004558116425847899,
      "loss": 5.3918,
      "mean_token_accuracy": 0.18263471573591233,
      "num_tokens": 48070224.0,
      "step": 23165
    },
    {
      "entropy": 5.715932321548462,
      "epoch": 2.10177793904209,
      "grad_norm": 2.125,
      "learning_rate": 0.0004557923834716596,
      "loss": 5.4106,
      "mean_token_accuracy": 0.17143329828977585,
      "num_tokens": 48080914.0,
      "step": 23170
    },
    {
      "entropy": 5.675404214859009,
      "epoch": 2.1022314949201744,
      "grad_norm": 2.78125,
      "learning_rate": 0.00045577312061969677,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17719782888889313,
      "num_tokens": 48090873.0,
      "step": 23175
    },
    {
      "entropy": 5.7836333274841305,
      "epoch": 2.102685050798258,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004557538540292997,
      "loss": 5.4327,
      "mean_token_accuracy": 0.17312687486410142,
      "num_tokens": 48101885.0,
      "step": 23180
    },
    {
      "entropy": 5.767099761962891,
      "epoch": 2.1031386066763424,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004557345837008669,
      "loss": 5.4133,
      "mean_token_accuracy": 0.17413544952869414,
      "num_tokens": 48111472.0,
      "step": 23185
    },
    {
      "entropy": 5.628871011734009,
      "epoch": 2.1035921625544267,
      "grad_norm": 2.6875,
      "learning_rate": 0.00045571530963479684,
      "loss": 5.2767,
      "mean_token_accuracy": 0.1817893862724304,
      "num_tokens": 48121351.0,
      "step": 23190
    },
    {
      "entropy": 5.738583183288574,
      "epoch": 2.104045718432511,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045569603183148814,
      "loss": 5.3586,
      "mean_token_accuracy": 0.1798534005880356,
      "num_tokens": 48131104.0,
      "step": 23195
    },
    {
      "entropy": 5.774147987365723,
      "epoch": 2.1044992743105952,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004556767502913395,
      "loss": 5.4707,
      "mean_token_accuracy": 0.1662584885954857,
      "num_tokens": 48142895.0,
      "step": 23200
    },
    {
      "entropy": 5.618874359130859,
      "epoch": 2.104952830188679,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045565746501474946,
      "loss": 5.2416,
      "mean_token_accuracy": 0.18075059801340104,
      "num_tokens": 48153294.0,
      "step": 23205
    },
    {
      "entropy": 5.731564140319824,
      "epoch": 2.1054063860667633,
      "grad_norm": 4.09375,
      "learning_rate": 0.00045563817600211714,
      "loss": 5.4273,
      "mean_token_accuracy": 0.17050953656435014,
      "num_tokens": 48164319.0,
      "step": 23210
    },
    {
      "entropy": 5.68043212890625,
      "epoch": 2.1058599419448476,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045561888325384124,
      "loss": 5.3812,
      "mean_token_accuracy": 0.1776420384645462,
      "num_tokens": 48173918.0,
      "step": 23215
    },
    {
      "entropy": 5.735940027236938,
      "epoch": 2.106313497822932,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004555995867703209,
      "loss": 5.3731,
      "mean_token_accuracy": 0.17378421723842621,
      "num_tokens": 48184127.0,
      "step": 23220
    },
    {
      "entropy": 5.678511762619019,
      "epoch": 2.106767053701016,
      "grad_norm": 2.9375,
      "learning_rate": 0.00045558028655195504,
      "loss": 5.3377,
      "mean_token_accuracy": 0.17100734114646912,
      "num_tokens": 48194198.0,
      "step": 23225
    },
    {
      "entropy": 5.647785758972168,
      "epoch": 2.1072206095791,
      "grad_norm": 3.453125,
      "learning_rate": 0.00045556098259914293,
      "loss": 5.3247,
      "mean_token_accuracy": 0.17657452076673508,
      "num_tokens": 48204654.0,
      "step": 23230
    },
    {
      "entropy": 5.717083501815796,
      "epoch": 2.1076741654571842,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045554167491228357,
      "loss": 5.4574,
      "mean_token_accuracy": 0.16625891923904418,
      "num_tokens": 48215322.0,
      "step": 23235
    },
    {
      "entropy": 5.667032909393311,
      "epoch": 2.1081277213352685,
      "grad_norm": 2.65625,
      "learning_rate": 0.00045552236349177637,
      "loss": 5.2708,
      "mean_token_accuracy": 0.18420125395059586,
      "num_tokens": 48226142.0,
      "step": 23240
    },
    {
      "entropy": 5.655144882202149,
      "epoch": 2.1085812772133528,
      "grad_norm": 3.234375,
      "learning_rate": 0.00045550304833802066,
      "loss": 5.2948,
      "mean_token_accuracy": 0.17575038820505143,
      "num_tokens": 48236896.0,
      "step": 23245
    },
    {
      "entropy": 5.584407663345337,
      "epoch": 2.109034833091437,
      "grad_norm": 2.953125,
      "learning_rate": 0.000455483729451416,
      "loss": 5.308,
      "mean_token_accuracy": 0.17780325263738633,
      "num_tokens": 48247425.0,
      "step": 23250
    },
    {
      "entropy": 5.731166934967041,
      "epoch": 2.109488388969521,
      "grad_norm": 3.15625,
      "learning_rate": 0.00045546440683236173,
      "loss": 5.4203,
      "mean_token_accuracy": 0.1720667749643326,
      "num_tokens": 48258560.0,
      "step": 23255
    },
    {
      "entropy": 5.754266023635864,
      "epoch": 2.109941944847605,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045544508048125763,
      "loss": 5.4544,
      "mean_token_accuracy": 0.17324440330266952,
      "num_tokens": 48269469.0,
      "step": 23260
    },
    {
      "entropy": 5.6901367664337155,
      "epoch": 2.1103955007256894,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004554257503985032,
      "loss": 5.3388,
      "mean_token_accuracy": 0.18047440946102142,
      "num_tokens": 48278847.0,
      "step": 23265
    },
    {
      "entropy": 5.6434258937835695,
      "epoch": 2.1108490566037736,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045540641658449834,
      "loss": 5.2813,
      "mean_token_accuracy": 0.17735383063554763,
      "num_tokens": 48288091.0,
      "step": 23270
    },
    {
      "entropy": 5.663916969299317,
      "epoch": 2.111302612481858,
      "grad_norm": 2.234375,
      "learning_rate": 0.00045538707903964275,
      "loss": 5.2747,
      "mean_token_accuracy": 0.17679276466369628,
      "num_tokens": 48297946.0,
      "step": 23275
    },
    {
      "entropy": 5.6722876071929935,
      "epoch": 2.1117561683599417,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045536773776433647,
      "loss": 5.3476,
      "mean_token_accuracy": 0.18119596540927888,
      "num_tokens": 48307650.0,
      "step": 23280
    },
    {
      "entropy": 5.733861351013184,
      "epoch": 2.112209724238026,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045534839275897937,
      "loss": 5.4831,
      "mean_token_accuracy": 0.16325066089630128,
      "num_tokens": 48318824.0,
      "step": 23285
    },
    {
      "entropy": 5.674610567092896,
      "epoch": 2.1126632801161103,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045532904402397157,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17629942893981934,
      "num_tokens": 48329909.0,
      "step": 23290
    },
    {
      "entropy": 5.71350998878479,
      "epoch": 2.1131168359941945,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004553096915597132,
      "loss": 5.4689,
      "mean_token_accuracy": 0.1689213767647743,
      "num_tokens": 48340142.0,
      "step": 23295
    },
    {
      "entropy": 5.699016332626343,
      "epoch": 2.113570391872279,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004552903353666044,
      "loss": 5.3183,
      "mean_token_accuracy": 0.1800950437784195,
      "num_tokens": 48350540.0,
      "step": 23300
    },
    {
      "entropy": 5.681776332855224,
      "epoch": 2.114023947750363,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045527097544504556,
      "loss": 5.2832,
      "mean_token_accuracy": 0.18349603712558746,
      "num_tokens": 48360153.0,
      "step": 23305
    },
    {
      "entropy": 5.613363981246948,
      "epoch": 2.114477503628447,
      "grad_norm": 2.765625,
      "learning_rate": 0.000455251611795437,
      "loss": 5.3282,
      "mean_token_accuracy": 0.17761099487543106,
      "num_tokens": 48370088.0,
      "step": 23310
    },
    {
      "entropy": 5.721817064285278,
      "epoch": 2.114931059506531,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004552322444181792,
      "loss": 5.3983,
      "mean_token_accuracy": 0.1727781727910042,
      "num_tokens": 48381325.0,
      "step": 23315
    },
    {
      "entropy": 5.721550464630127,
      "epoch": 2.1153846153846154,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004552128733136726,
      "loss": 5.3618,
      "mean_token_accuracy": 0.173321533203125,
      "num_tokens": 48393077.0,
      "step": 23320
    },
    {
      "entropy": 5.685474300384522,
      "epoch": 2.1158381712626997,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004551934984823179,
      "loss": 5.3671,
      "mean_token_accuracy": 0.17693899124860762,
      "num_tokens": 48403260.0,
      "step": 23325
    },
    {
      "entropy": 5.6789155960083,
      "epoch": 2.116291727140784,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004551741199245157,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17964747846126555,
      "num_tokens": 48413616.0,
      "step": 23330
    },
    {
      "entropy": 5.709835243225098,
      "epoch": 2.1167452830188678,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045515473764066685,
      "loss": 5.3499,
      "mean_token_accuracy": 0.17668165564537047,
      "num_tokens": 48424189.0,
      "step": 23335
    },
    {
      "entropy": 5.650756597518921,
      "epoch": 2.117198838896952,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045513535163117205,
      "loss": 5.4155,
      "mean_token_accuracy": 0.17464146614074708,
      "num_tokens": 48433603.0,
      "step": 23340
    },
    {
      "entropy": 5.789517021179199,
      "epoch": 2.1176523947750363,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004551159618964323,
      "loss": 5.3866,
      "mean_token_accuracy": 0.17448541522026062,
      "num_tokens": 48443043.0,
      "step": 23345
    },
    {
      "entropy": 5.685244369506836,
      "epoch": 2.1181059506531206,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004550965684368486,
      "loss": 5.2761,
      "mean_token_accuracy": 0.19112497568130493,
      "num_tokens": 48452765.0,
      "step": 23350
    },
    {
      "entropy": 5.650390338897705,
      "epoch": 2.118559506531205,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004550771712528219,
      "loss": 5.3339,
      "mean_token_accuracy": 0.1739571437239647,
      "num_tokens": 48463191.0,
      "step": 23355
    },
    {
      "entropy": 5.663244390487671,
      "epoch": 2.1190130624092887,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004550577703447534,
      "loss": 5.3393,
      "mean_token_accuracy": 0.17711763828992844,
      "num_tokens": 48473280.0,
      "step": 23360
    },
    {
      "entropy": 5.701294040679931,
      "epoch": 2.119466618287373,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045503836571304443,
      "loss": 5.288,
      "mean_token_accuracy": 0.17841034829616548,
      "num_tokens": 48483465.0,
      "step": 23365
    },
    {
      "entropy": 5.670509624481201,
      "epoch": 2.119920174165457,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004550189573580961,
      "loss": 5.2861,
      "mean_token_accuracy": 0.18078475445508957,
      "num_tokens": 48494497.0,
      "step": 23370
    },
    {
      "entropy": 5.689207410812378,
      "epoch": 2.1203737300435415,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004549995452803098,
      "loss": 5.3744,
      "mean_token_accuracy": 0.17860814034938813,
      "num_tokens": 48504106.0,
      "step": 23375
    },
    {
      "entropy": 5.803400182723999,
      "epoch": 2.1208272859216257,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004549801294800872,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1694880411028862,
      "num_tokens": 48515729.0,
      "step": 23380
    },
    {
      "entropy": 5.723779487609863,
      "epoch": 2.1212808417997095,
      "grad_norm": 2.75,
      "learning_rate": 0.0004549607099578295,
      "loss": 5.4366,
      "mean_token_accuracy": 0.16740094870328903,
      "num_tokens": 48526611.0,
      "step": 23385
    },
    {
      "entropy": 5.614547157287598,
      "epoch": 2.121734397677794,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004549412867139386,
      "loss": 5.3011,
      "mean_token_accuracy": 0.17402708083391188,
      "num_tokens": 48537016.0,
      "step": 23390
    },
    {
      "entropy": 5.719745731353759,
      "epoch": 2.122187953555878,
      "grad_norm": 2.625,
      "learning_rate": 0.00045492185974881587,
      "loss": 5.3927,
      "mean_token_accuracy": 0.17100141197443008,
      "num_tokens": 48546786.0,
      "step": 23395
    },
    {
      "entropy": 5.728090715408325,
      "epoch": 2.1226415094339623,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004549024290628634,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17493391484022142,
      "num_tokens": 48556306.0,
      "step": 23400
    },
    {
      "entropy": 5.744469213485718,
      "epoch": 2.1230950653120466,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004548829946564827,
      "loss": 5.431,
      "mean_token_accuracy": 0.1744708999991417,
      "num_tokens": 48566303.0,
      "step": 23405
    },
    {
      "entropy": 5.7656834602355955,
      "epoch": 2.1235486211901304,
      "grad_norm": 2.0,
      "learning_rate": 0.0004548635565300759,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17793896049261093,
      "num_tokens": 48576313.0,
      "step": 23410
    },
    {
      "entropy": 5.6690099239349365,
      "epoch": 2.1240021770682147,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004548441146840449,
      "loss": 5.3539,
      "mean_token_accuracy": 0.1781396359205246,
      "num_tokens": 48587133.0,
      "step": 23415
    },
    {
      "entropy": 5.774262952804565,
      "epoch": 2.124455732946299,
      "grad_norm": 2.921875,
      "learning_rate": 0.00045482466911879184,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16961318552494048,
      "num_tokens": 48598524.0,
      "step": 23420
    },
    {
      "entropy": 5.699971628189087,
      "epoch": 2.1249092888243832,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045480521983471875,
      "loss": 5.3332,
      "mean_token_accuracy": 0.17774871736764908,
      "num_tokens": 48609870.0,
      "step": 23425
    },
    {
      "entropy": 5.711522245407105,
      "epoch": 2.1253628447024675,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004547857668322279,
      "loss": 5.4018,
      "mean_token_accuracy": 0.17904150784015654,
      "num_tokens": 48619828.0,
      "step": 23430
    },
    {
      "entropy": 5.654345941543579,
      "epoch": 2.1258164005805513,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004547663101117215,
      "loss": 5.3094,
      "mean_token_accuracy": 0.18113369047641753,
      "num_tokens": 48630361.0,
      "step": 23435
    },
    {
      "entropy": 5.683588123321533,
      "epoch": 2.1262699564586356,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045474684967360216,
      "loss": 5.3529,
      "mean_token_accuracy": 0.1701095223426819,
      "num_tokens": 48641113.0,
      "step": 23440
    },
    {
      "entropy": 5.782934474945068,
      "epoch": 2.12672351233672,
      "grad_norm": 1.859375,
      "learning_rate": 0.000454727385518272,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16320455223321914,
      "num_tokens": 48651777.0,
      "step": 23445
    },
    {
      "entropy": 5.793789005279541,
      "epoch": 2.127177068214804,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004547079176461338,
      "loss": 5.465,
      "mean_token_accuracy": 0.1685747116804123,
      "num_tokens": 48663237.0,
      "step": 23450
    },
    {
      "entropy": 5.687932634353638,
      "epoch": 2.1276306240928884,
      "grad_norm": 2.125,
      "learning_rate": 0.0004546884460575901,
      "loss": 5.3519,
      "mean_token_accuracy": 0.17945593297481538,
      "num_tokens": 48673932.0,
      "step": 23455
    },
    {
      "entropy": 5.646729040145874,
      "epoch": 2.1280841799709727,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004546689707530434,
      "loss": 5.2932,
      "mean_token_accuracy": 0.18333759754896164,
      "num_tokens": 48684304.0,
      "step": 23460
    },
    {
      "entropy": 5.646184968948364,
      "epoch": 2.1285377358490565,
      "grad_norm": 1.921875,
      "learning_rate": 0.00045464949173289676,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17788189500570298,
      "num_tokens": 48695569.0,
      "step": 23465
    },
    {
      "entropy": 5.6668284893035885,
      "epoch": 2.1289912917271407,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004546300089975528,
      "loss": 5.3573,
      "mean_token_accuracy": 0.17756066024303435,
      "num_tokens": 48706550.0,
      "step": 23470
    },
    {
      "entropy": 5.737710094451904,
      "epoch": 2.129444847605225,
      "grad_norm": 1.875,
      "learning_rate": 0.00045461052254741445,
      "loss": 5.4044,
      "mean_token_accuracy": 0.17049935460090637,
      "num_tokens": 48716852.0,
      "step": 23475
    },
    {
      "entropy": 5.626887559890747,
      "epoch": 2.1298984034833093,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045459103238288474,
      "loss": 5.3053,
      "mean_token_accuracy": 0.17556218057870865,
      "num_tokens": 48726607.0,
      "step": 23480
    },
    {
      "entropy": 5.701097917556763,
      "epoch": 2.130351959361393,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004545715385043667,
      "loss": 5.5217,
      "mean_token_accuracy": 0.16378091126680375,
      "num_tokens": 48738011.0,
      "step": 23485
    },
    {
      "entropy": 5.764771032333374,
      "epoch": 2.1308055152394774,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004545520409122635,
      "loss": 5.416,
      "mean_token_accuracy": 0.1644475966691971,
      "num_tokens": 48747899.0,
      "step": 23490
    },
    {
      "entropy": 5.739223480224609,
      "epoch": 2.1312590711175616,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004545325396069783,
      "loss": 5.4182,
      "mean_token_accuracy": 0.17154274731874466,
      "num_tokens": 48758719.0,
      "step": 23495
    },
    {
      "entropy": 5.77618522644043,
      "epoch": 2.131712626995646,
      "grad_norm": 2.25,
      "learning_rate": 0.0004545130345889145,
      "loss": 5.4387,
      "mean_token_accuracy": 0.17486509680747986,
      "num_tokens": 48769677.0,
      "step": 23500
    },
    {
      "entropy": 5.672484922409057,
      "epoch": 2.13216618287373,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004544935258584754,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17242437303066255,
      "num_tokens": 48778440.0,
      "step": 23505
    },
    {
      "entropy": 5.752479457855225,
      "epoch": 2.1326197387518144,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004544740134160644,
      "loss": 5.464,
      "mean_token_accuracy": 0.17118218094110488,
      "num_tokens": 48788328.0,
      "step": 23510
    },
    {
      "entropy": 5.720599794387818,
      "epoch": 2.1330732946298983,
      "grad_norm": 2.125,
      "learning_rate": 0.0004544544972620851,
      "loss": 5.4172,
      "mean_token_accuracy": 0.16956216096878052,
      "num_tokens": 48799218.0,
      "step": 23515
    },
    {
      "entropy": 5.718430376052856,
      "epoch": 2.1335268505079825,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004544349773969411,
      "loss": 5.367,
      "mean_token_accuracy": 0.16887849494814872,
      "num_tokens": 48809648.0,
      "step": 23520
    },
    {
      "entropy": 5.725247049331665,
      "epoch": 2.133980406386067,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004544154538210359,
      "loss": 5.3885,
      "mean_token_accuracy": 0.17086625695228577,
      "num_tokens": 48819965.0,
      "step": 23525
    },
    {
      "entropy": 5.799205017089844,
      "epoch": 2.134433962264151,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004543959265347735,
      "loss": 5.4359,
      "mean_token_accuracy": 0.17584224343299865,
      "num_tokens": 48829862.0,
      "step": 23530
    },
    {
      "entropy": 5.735884141921997,
      "epoch": 2.1348875181422353,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045437639553855757,
      "loss": 5.3783,
      "mean_token_accuracy": 0.17146483063697815,
      "num_tokens": 48840551.0,
      "step": 23535
    },
    {
      "entropy": 5.696857881546021,
      "epoch": 2.135341074020319,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004543568608327921,
      "loss": 5.3573,
      "mean_token_accuracy": 0.1828202560544014,
      "num_tokens": 48851032.0,
      "step": 23540
    },
    {
      "entropy": 5.695530796051026,
      "epoch": 2.1357946298984034,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045433732241788105,
      "loss": 5.4249,
      "mean_token_accuracy": 0.1731689378619194,
      "num_tokens": 48861773.0,
      "step": 23545
    },
    {
      "entropy": 5.709213542938232,
      "epoch": 2.1362481857764877,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045431778029422846,
      "loss": 5.4641,
      "mean_token_accuracy": 0.171664035320282,
      "num_tokens": 48872430.0,
      "step": 23550
    },
    {
      "entropy": 5.681395101547241,
      "epoch": 2.136701741654572,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004542982344622385,
      "loss": 5.2716,
      "mean_token_accuracy": 0.17576860785484313,
      "num_tokens": 48882575.0,
      "step": 23555
    },
    {
      "entropy": 5.647830533981323,
      "epoch": 2.137155297532656,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045427868492231533,
      "loss": 5.3561,
      "mean_token_accuracy": 0.17409990280866622,
      "num_tokens": 48893337.0,
      "step": 23560
    },
    {
      "entropy": 5.801657485961914,
      "epoch": 2.13760885341074,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045425913167486333,
      "loss": 5.5106,
      "mean_token_accuracy": 0.17068567126989365,
      "num_tokens": 48904675.0,
      "step": 23565
    },
    {
      "entropy": 5.724964570999146,
      "epoch": 2.1380624092888243,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045423957472028677,
      "loss": 5.3062,
      "mean_token_accuracy": 0.18483853340148926,
      "num_tokens": 48914831.0,
      "step": 23570
    },
    {
      "entropy": 5.628698682785034,
      "epoch": 2.1385159651669086,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004542200140589901,
      "loss": 5.2751,
      "mean_token_accuracy": 0.17937944680452347,
      "num_tokens": 48924574.0,
      "step": 23575
    },
    {
      "entropy": 5.6436827182769775,
      "epoch": 2.138969521044993,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045420044969137787,
      "loss": 5.4147,
      "mean_token_accuracy": 0.1758163571357727,
      "num_tokens": 48933921.0,
      "step": 23580
    },
    {
      "entropy": 5.677951669692993,
      "epoch": 2.139423076923077,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004541808816178547,
      "loss": 5.3738,
      "mean_token_accuracy": 0.17326900660991668,
      "num_tokens": 48944571.0,
      "step": 23585
    },
    {
      "entropy": 5.700808525085449,
      "epoch": 2.139876632801161,
      "grad_norm": 5.0,
      "learning_rate": 0.00045416130983882524,
      "loss": 5.399,
      "mean_token_accuracy": 0.17478104382753373,
      "num_tokens": 48954569.0,
      "step": 23590
    },
    {
      "entropy": 5.704714393615722,
      "epoch": 2.140330188679245,
      "grad_norm": 2.625,
      "learning_rate": 0.00045414173435469427,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17240772247314454,
      "num_tokens": 48965886.0,
      "step": 23595
    },
    {
      "entropy": 5.732536458969117,
      "epoch": 2.1407837445573294,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045412215516586654,
      "loss": 5.3788,
      "mean_token_accuracy": 0.1740315079689026,
      "num_tokens": 48976219.0,
      "step": 23600
    },
    {
      "entropy": 5.641070795059204,
      "epoch": 2.1412373004354137,
      "grad_norm": 4.15625,
      "learning_rate": 0.000454102572272747,
      "loss": 5.3431,
      "mean_token_accuracy": 0.17349760681390763,
      "num_tokens": 48986322.0,
      "step": 23605
    },
    {
      "entropy": 5.635484266281128,
      "epoch": 2.141690856313498,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004540829856757406,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1758939191699028,
      "num_tokens": 48996297.0,
      "step": 23610
    },
    {
      "entropy": 5.648513746261597,
      "epoch": 2.142144412191582,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004540633953752525,
      "loss": 5.3645,
      "mean_token_accuracy": 0.17998359203338624,
      "num_tokens": 49006165.0,
      "step": 23615
    },
    {
      "entropy": 5.706613302230835,
      "epoch": 2.142597968069666,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004540438013716876,
      "loss": 5.3175,
      "mean_token_accuracy": 0.16563782393932341,
      "num_tokens": 49017278.0,
      "step": 23620
    },
    {
      "entropy": 5.598768138885498,
      "epoch": 2.1430515239477503,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004540242036654514,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17224858701229095,
      "num_tokens": 49027088.0,
      "step": 23625
    },
    {
      "entropy": 5.675223779678345,
      "epoch": 2.1435050798258346,
      "grad_norm": 2.34375,
      "learning_rate": 0.000454004602256949,
      "loss": 5.2747,
      "mean_token_accuracy": 0.17645884454250335,
      "num_tokens": 49038156.0,
      "step": 23630
    },
    {
      "entropy": 5.72030816078186,
      "epoch": 2.143958635703919,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004539849971465858,
      "loss": 5.364,
      "mean_token_accuracy": 0.17651452571153642,
      "num_tokens": 49047480.0,
      "step": 23635
    },
    {
      "entropy": 5.6858008861541744,
      "epoch": 2.1444121915820027,
      "grad_norm": 3.25,
      "learning_rate": 0.00045396538833476725,
      "loss": 5.4076,
      "mean_token_accuracy": 0.17420943528413774,
      "num_tokens": 49057373.0,
      "step": 23640
    },
    {
      "entropy": 5.692072772979737,
      "epoch": 2.144865747460087,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004539457758218989,
      "loss": 5.3877,
      "mean_token_accuracy": 0.1784567043185234,
      "num_tokens": 49068380.0,
      "step": 23645
    },
    {
      "entropy": 5.713856744766235,
      "epoch": 2.1453193033381712,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004539261596083863,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17510260939598082,
      "num_tokens": 49078646.0,
      "step": 23650
    },
    {
      "entropy": 5.650618648529052,
      "epoch": 2.1457728592162555,
      "grad_norm": 2.875,
      "learning_rate": 0.0004539065396946351,
      "loss": 5.304,
      "mean_token_accuracy": 0.18309558779001237,
      "num_tokens": 49089281.0,
      "step": 23655
    },
    {
      "entropy": 5.691441583633423,
      "epoch": 2.1462264150943398,
      "grad_norm": 2.0,
      "learning_rate": 0.0004538869160810511,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17201095521450044,
      "num_tokens": 49099514.0,
      "step": 23660
    },
    {
      "entropy": 5.741053199768066,
      "epoch": 2.146679970972424,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045386728876804,
      "loss": 5.4242,
      "mean_token_accuracy": 0.17366275787353516,
      "num_tokens": 49110247.0,
      "step": 23665
    },
    {
      "entropy": 5.7587635040283205,
      "epoch": 2.147133526850508,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045384765775600787,
      "loss": 5.47,
      "mean_token_accuracy": 0.16695613265037537,
      "num_tokens": 49122492.0,
      "step": 23670
    },
    {
      "entropy": 5.72358775138855,
      "epoch": 2.147587082728592,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004538280230453606,
      "loss": 5.3605,
      "mean_token_accuracy": 0.1769393801689148,
      "num_tokens": 49132395.0,
      "step": 23675
    },
    {
      "entropy": 5.750011348724366,
      "epoch": 2.1480406386066764,
      "grad_norm": 2.75,
      "learning_rate": 0.0004538083846365042,
      "loss": 5.3893,
      "mean_token_accuracy": 0.17693014293909073,
      "num_tokens": 49142909.0,
      "step": 23680
    },
    {
      "entropy": 5.662167310714722,
      "epoch": 2.1484941944847606,
      "grad_norm": 2.25,
      "learning_rate": 0.0004537887425298448,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17071327418088914,
      "num_tokens": 49152933.0,
      "step": 23685
    },
    {
      "entropy": 5.646643114089966,
      "epoch": 2.1489477503628445,
      "grad_norm": 3.609375,
      "learning_rate": 0.00045376909672578875,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17736134827136993,
      "num_tokens": 49163595.0,
      "step": 23690
    },
    {
      "entropy": 5.664646339416504,
      "epoch": 2.1494013062409287,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004537494472247422,
      "loss": 5.3401,
      "mean_token_accuracy": 0.17659894824028016,
      "num_tokens": 49174664.0,
      "step": 23695
    },
    {
      "entropy": 5.746810531616211,
      "epoch": 2.149854862119013,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004537297940271115,
      "loss": 5.3987,
      "mean_token_accuracy": 0.176193967461586,
      "num_tokens": 49185105.0,
      "step": 23700
    },
    {
      "entropy": 5.631782054901123,
      "epoch": 2.1503084179970973,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004537101371333031,
      "loss": 5.3072,
      "mean_token_accuracy": 0.17547917515039443,
      "num_tokens": 49194679.0,
      "step": 23705
    },
    {
      "entropy": 5.623754692077637,
      "epoch": 2.1507619738751815,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045369047654372356,
      "loss": 5.269,
      "mean_token_accuracy": 0.17535709887742995,
      "num_tokens": 49204563.0,
      "step": 23710
    },
    {
      "entropy": 5.705962610244751,
      "epoch": 2.151215529753266,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004536708122587795,
      "loss": 5.3364,
      "mean_token_accuracy": 0.1805448576807976,
      "num_tokens": 49213759.0,
      "step": 23715
    },
    {
      "entropy": 5.624365568161011,
      "epoch": 2.1516690856313496,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004536511442788774,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18411164432764054,
      "num_tokens": 49223604.0,
      "step": 23720
    },
    {
      "entropy": 5.682430458068848,
      "epoch": 2.152122641509434,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004536314726044241,
      "loss": 5.3787,
      "mean_token_accuracy": 0.17334912866353988,
      "num_tokens": 49234000.0,
      "step": 23725
    },
    {
      "entropy": 5.7347780704498295,
      "epoch": 2.152576197387518,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045361179723582654,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17824837565422058,
      "num_tokens": 49244874.0,
      "step": 23730
    },
    {
      "entropy": 5.7541906356811525,
      "epoch": 2.1530297532656024,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004535921181734914,
      "loss": 5.5379,
      "mean_token_accuracy": 0.16916103065013885,
      "num_tokens": 49255495.0,
      "step": 23735
    },
    {
      "entropy": 5.679804277420044,
      "epoch": 2.1534833091436867,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004535724354178257,
      "loss": 5.3015,
      "mean_token_accuracy": 0.17692258059978486,
      "num_tokens": 49265357.0,
      "step": 23740
    },
    {
      "entropy": 5.738633251190185,
      "epoch": 2.1539368650217705,
      "grad_norm": 2.5,
      "learning_rate": 0.00045355274896923664,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16826530396938325,
      "num_tokens": 49276396.0,
      "step": 23745
    },
    {
      "entropy": 5.653690528869629,
      "epoch": 2.1543904208998548,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004535330588281312,
      "loss": 5.2618,
      "mean_token_accuracy": 0.18154412657022476,
      "num_tokens": 49285925.0,
      "step": 23750
    },
    {
      "entropy": 5.586862421035766,
      "epoch": 2.154843976777939,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004535133649949166,
      "loss": 5.3571,
      "mean_token_accuracy": 0.18121253103017806,
      "num_tokens": 49296493.0,
      "step": 23755
    },
    {
      "entropy": 5.671215057373047,
      "epoch": 2.1552975326560233,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004534936674700001,
      "loss": 5.3196,
      "mean_token_accuracy": 0.17983419299125672,
      "num_tokens": 49307273.0,
      "step": 23760
    },
    {
      "entropy": 5.644394826889038,
      "epoch": 2.1557510885341076,
      "grad_norm": 2.65625,
      "learning_rate": 0.00045347396625378914,
      "loss": 5.2982,
      "mean_token_accuracy": 0.18326902836561204,
      "num_tokens": 49317055.0,
      "step": 23765
    },
    {
      "entropy": 5.647957611083984,
      "epoch": 2.1562046444121914,
      "grad_norm": 2.609375,
      "learning_rate": 0.00045345426134669103,
      "loss": 5.3513,
      "mean_token_accuracy": 0.1723825141787529,
      "num_tokens": 49327420.0,
      "step": 23770
    },
    {
      "entropy": 5.666400051116943,
      "epoch": 2.1566582002902757,
      "grad_norm": 4.90625,
      "learning_rate": 0.00045343455274911325,
      "loss": 5.4482,
      "mean_token_accuracy": 0.1685351237654686,
      "num_tokens": 49338323.0,
      "step": 23775
    },
    {
      "entropy": 5.704577016830444,
      "epoch": 2.15711175616836,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004534148404614635,
      "loss": 5.4099,
      "mean_token_accuracy": 0.17430777549743653,
      "num_tokens": 49348364.0,
      "step": 23780
    },
    {
      "entropy": 5.781159257888794,
      "epoch": 2.157565312046444,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045339512448414943,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17542660981416702,
      "num_tokens": 49358274.0,
      "step": 23785
    },
    {
      "entropy": 5.63041672706604,
      "epoch": 2.1580188679245285,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004533754048175787,
      "loss": 5.2737,
      "mean_token_accuracy": 0.17383868396282195,
      "num_tokens": 49368651.0,
      "step": 23790
    },
    {
      "entropy": 5.622497081756592,
      "epoch": 2.1584724238026123,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004533556814621591,
      "loss": 5.2949,
      "mean_token_accuracy": 0.173082435131073,
      "num_tokens": 49378363.0,
      "step": 23795
    },
    {
      "entropy": 5.657915830612183,
      "epoch": 2.1589259796806965,
      "grad_norm": 2.984375,
      "learning_rate": 0.00045333595441829865,
      "loss": 5.3281,
      "mean_token_accuracy": 0.17905047088861464,
      "num_tokens": 49388241.0,
      "step": 23800
    },
    {
      "entropy": 5.632946586608886,
      "epoch": 2.159379535558781,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004533162236864051,
      "loss": 5.3132,
      "mean_token_accuracy": 0.17508240938186645,
      "num_tokens": 49397678.0,
      "step": 23805
    },
    {
      "entropy": 5.6932470321655275,
      "epoch": 2.159833091436865,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004532964892668867,
      "loss": 5.4301,
      "mean_token_accuracy": 0.17176597490906714,
      "num_tokens": 49407368.0,
      "step": 23810
    },
    {
      "entropy": 5.703235197067261,
      "epoch": 2.1602866473149493,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045327675116015146,
      "loss": 5.357,
      "mean_token_accuracy": 0.17703022062778473,
      "num_tokens": 49417926.0,
      "step": 23815
    },
    {
      "entropy": 5.760638809204101,
      "epoch": 2.1607402031930336,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004532570093666075,
      "loss": 5.4401,
      "mean_token_accuracy": 0.16397894471883773,
      "num_tokens": 49427899.0,
      "step": 23820
    },
    {
      "entropy": 5.683732175827027,
      "epoch": 2.1611937590711174,
      "grad_norm": 2.0,
      "learning_rate": 0.0004532372638866633,
      "loss": 5.3246,
      "mean_token_accuracy": 0.18212025612592697,
      "num_tokens": 49439102.0,
      "step": 23825
    },
    {
      "entropy": 5.635964202880859,
      "epoch": 2.1616473149492017,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00045321751472072693,
      "loss": 5.2595,
      "mean_token_accuracy": 0.18512066453695297,
      "num_tokens": 49448013.0,
      "step": 23830
    },
    {
      "entropy": 5.558284091949463,
      "epoch": 2.162100870827286,
      "grad_norm": 1.640625,
      "learning_rate": 0.00045319776186920695,
      "loss": 5.3198,
      "mean_token_accuracy": 0.18118231892585754,
      "num_tokens": 49457620.0,
      "step": 23835
    },
    {
      "entropy": 5.654453182220459,
      "epoch": 2.1625544267053702,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045317800533251193,
      "loss": 5.3419,
      "mean_token_accuracy": 0.172549244761467,
      "num_tokens": 49468038.0,
      "step": 23840
    },
    {
      "entropy": 5.626783227920532,
      "epoch": 2.163007982583454,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045315824511105027,
      "loss": 5.2554,
      "mean_token_accuracy": 0.1792191445827484,
      "num_tokens": 49477714.0,
      "step": 23845
    },
    {
      "entropy": 5.71697096824646,
      "epoch": 2.1634615384615383,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045313848120523073,
      "loss": 5.2976,
      "mean_token_accuracy": 0.17661184519529344,
      "num_tokens": 49488214.0,
      "step": 23850
    },
    {
      "entropy": 5.672116136550903,
      "epoch": 2.1639150943396226,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045311871361546205,
      "loss": 5.4228,
      "mean_token_accuracy": 0.17845352590084076,
      "num_tokens": 49498444.0,
      "step": 23855
    },
    {
      "entropy": 5.659992218017578,
      "epoch": 2.164368650217707,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004530989423421529,
      "loss": 5.2687,
      "mean_token_accuracy": 0.18276670426130295,
      "num_tokens": 49508759.0,
      "step": 23860
    },
    {
      "entropy": 5.6174708843231205,
      "epoch": 2.164822206095791,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045307916738571223,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18365022242069245,
      "num_tokens": 49518787.0,
      "step": 23865
    },
    {
      "entropy": 5.603063535690308,
      "epoch": 2.1652757619738754,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004530593887465491,
      "loss": 5.3263,
      "mean_token_accuracy": 0.18313729614019394,
      "num_tokens": 49528130.0,
      "step": 23870
    },
    {
      "entropy": 5.766613483428955,
      "epoch": 2.165729317851959,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045303960642507235,
      "loss": 5.5017,
      "mean_token_accuracy": 0.1701413094997406,
      "num_tokens": 49539128.0,
      "step": 23875
    },
    {
      "entropy": 5.641251754760742,
      "epoch": 2.1661828737300435,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045301982042169123,
      "loss": 5.3667,
      "mean_token_accuracy": 0.17658547908067704,
      "num_tokens": 49550571.0,
      "step": 23880
    },
    {
      "entropy": 5.654483985900879,
      "epoch": 2.1666364296081277,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045300003073681485,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17705845981836318,
      "num_tokens": 49561106.0,
      "step": 23885
    },
    {
      "entropy": 5.645529794692993,
      "epoch": 2.167089985486212,
      "grad_norm": 2.25,
      "learning_rate": 0.00045298023737085243,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1712099254131317,
      "num_tokens": 49571389.0,
      "step": 23890
    },
    {
      "entropy": 5.647616195678711,
      "epoch": 2.1675435413642963,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004529604403242134,
      "loss": 5.3621,
      "mean_token_accuracy": 0.1746892973780632,
      "num_tokens": 49582224.0,
      "step": 23895
    },
    {
      "entropy": 5.666357946395874,
      "epoch": 2.16799709724238,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045294063959730707,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17149990797042847,
      "num_tokens": 49592413.0,
      "step": 23900
    },
    {
      "entropy": 5.731571435928345,
      "epoch": 2.1684506531204644,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045292083519054297,
      "loss": 5.4377,
      "mean_token_accuracy": 0.16184748262166976,
      "num_tokens": 49602796.0,
      "step": 23905
    },
    {
      "entropy": 5.691939973831177,
      "epoch": 2.1689042089985486,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004529010271043307,
      "loss": 5.3273,
      "mean_token_accuracy": 0.17833031117916107,
      "num_tokens": 49613625.0,
      "step": 23910
    },
    {
      "entropy": 5.785305547714233,
      "epoch": 2.169357764876633,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045288121533907977,
      "loss": 5.469,
      "mean_token_accuracy": 0.17159054428339005,
      "num_tokens": 49624719.0,
      "step": 23915
    },
    {
      "entropy": 5.709560775756836,
      "epoch": 2.169811320754717,
      "grad_norm": 2.0,
      "learning_rate": 0.00045286139989520006,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1751367226243019,
      "num_tokens": 49634916.0,
      "step": 23920
    },
    {
      "entropy": 5.6497115135192875,
      "epoch": 2.170264876632801,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004528415807731012,
      "loss": 5.2957,
      "mean_token_accuracy": 0.1784551426768303,
      "num_tokens": 49645215.0,
      "step": 23925
    },
    {
      "entropy": 5.674240446090698,
      "epoch": 2.1707184325108853,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045282175797319317,
      "loss": 5.3853,
      "mean_token_accuracy": 0.17304304540157317,
      "num_tokens": 49655805.0,
      "step": 23930
    },
    {
      "entropy": 5.786383962631225,
      "epoch": 2.1711719883889695,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004528019314958859,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17403020411729814,
      "num_tokens": 49665264.0,
      "step": 23935
    },
    {
      "entropy": 5.7248814582824705,
      "epoch": 2.171625544267054,
      "grad_norm": 2.890625,
      "learning_rate": 0.00045278210134158935,
      "loss": 5.4355,
      "mean_token_accuracy": 0.16671450287103654,
      "num_tokens": 49676152.0,
      "step": 23940
    },
    {
      "entropy": 5.664438486099243,
      "epoch": 2.172079100145138,
      "grad_norm": 1.625,
      "learning_rate": 0.0004527622675107137,
      "loss": 5.3456,
      "mean_token_accuracy": 0.17083630561828614,
      "num_tokens": 49686762.0,
      "step": 23945
    },
    {
      "entropy": 5.709523773193359,
      "epoch": 2.172532656023222,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045274243000366897,
      "loss": 5.4064,
      "mean_token_accuracy": 0.17526543140411377,
      "num_tokens": 49697504.0,
      "step": 23950
    },
    {
      "entropy": 5.6673877239227295,
      "epoch": 2.172986211901306,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004527225888208655,
      "loss": 5.4333,
      "mean_token_accuracy": 0.16775091886520385,
      "num_tokens": 49709197.0,
      "step": 23955
    },
    {
      "entropy": 5.702869844436646,
      "epoch": 2.1734397677793904,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004527027439627137,
      "loss": 5.3894,
      "mean_token_accuracy": 0.17140923589468002,
      "num_tokens": 49719764.0,
      "step": 23960
    },
    {
      "entropy": 5.712185192108154,
      "epoch": 2.1738933236574747,
      "grad_norm": 1.734375,
      "learning_rate": 0.00045268289542962377,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17184640169143678,
      "num_tokens": 49730363.0,
      "step": 23965
    },
    {
      "entropy": 5.674692440032959,
      "epoch": 2.174346879535559,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004526630432220063,
      "loss": 5.3776,
      "mean_token_accuracy": 0.16697140038013458,
      "num_tokens": 49741037.0,
      "step": 23970
    },
    {
      "entropy": 5.6523932933807375,
      "epoch": 2.1748004354136428,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004526431873402719,
      "loss": 5.2969,
      "mean_token_accuracy": 0.18330187499523162,
      "num_tokens": 49750127.0,
      "step": 23975
    },
    {
      "entropy": 5.768195676803589,
      "epoch": 2.175253991291727,
      "grad_norm": 1.96875,
      "learning_rate": 0.000452623327784831,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1629353679716587,
      "num_tokens": 49760298.0,
      "step": 23980
    },
    {
      "entropy": 5.77179970741272,
      "epoch": 2.1757075471698113,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004526034645560945,
      "loss": 5.3901,
      "mean_token_accuracy": 0.17297538369894028,
      "num_tokens": 49770841.0,
      "step": 23985
    },
    {
      "entropy": 5.64270281791687,
      "epoch": 2.1761611030478956,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045258359765447306,
      "loss": 5.3642,
      "mean_token_accuracy": 0.17446827590465547,
      "num_tokens": 49781000.0,
      "step": 23990
    },
    {
      "entropy": 5.729996824264527,
      "epoch": 2.17661465892598,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004525637270803776,
      "loss": 5.3759,
      "mean_token_accuracy": 0.1712614804506302,
      "num_tokens": 49790886.0,
      "step": 23995
    },
    {
      "entropy": 5.625037813186646,
      "epoch": 2.1770682148040637,
      "grad_norm": 2.25,
      "learning_rate": 0.00045254385283421896,
      "loss": 5.3702,
      "mean_token_accuracy": 0.18001581877470016,
      "num_tokens": 49801477.0,
      "step": 24000
    },
    {
      "epoch": 2.1770682148040637,
      "eval_entropy": 5.3800982762246115,
      "eval_loss": 5.457826614379883,
      "eval_mean_token_accuracy": 0.17901436503875445,
      "eval_num_tokens": 49801477.0,
      "eval_runtime": 26.8078,
      "eval_samples_per_second": 1319.017,
      "eval_steps_per_second": 164.877,
      "step": 24000
    },
    {
      "entropy": 5.660412311553955,
      "epoch": 2.177521770682148,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004525239749164082,
      "loss": 5.4249,
      "mean_token_accuracy": 0.17428149282932281,
      "num_tokens": 49812044.0,
      "step": 24005
    },
    {
      "entropy": 5.688362312316895,
      "epoch": 2.177975326560232,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004525040933273564,
      "loss": 5.2936,
      "mean_token_accuracy": 0.17787100672721862,
      "num_tokens": 49821559.0,
      "step": 24010
    },
    {
      "entropy": 5.756610345840454,
      "epoch": 2.1784288824383164,
      "grad_norm": 2.984375,
      "learning_rate": 0.00045248420806747483,
      "loss": 5.438,
      "mean_token_accuracy": 0.1690705105662346,
      "num_tokens": 49832937.0,
      "step": 24015
    },
    {
      "entropy": 5.711782169342041,
      "epoch": 2.1788824383164007,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004524643191371745,
      "loss": 5.3749,
      "mean_token_accuracy": 0.1747639536857605,
      "num_tokens": 49843176.0,
      "step": 24020
    },
    {
      "entropy": 5.651652526855469,
      "epoch": 2.179335994194485,
      "grad_norm": 2.25,
      "learning_rate": 0.00045244442653686684,
      "loss": 5.2952,
      "mean_token_accuracy": 0.18792311400175093,
      "num_tokens": 49852522.0,
      "step": 24025
    },
    {
      "entropy": 5.622285509109497,
      "epoch": 2.179789550072569,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045242453026696325,
      "loss": 5.3097,
      "mean_token_accuracy": 0.18044510632753372,
      "num_tokens": 49863143.0,
      "step": 24030
    },
    {
      "entropy": 5.75468430519104,
      "epoch": 2.180243105950653,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004524046303278752,
      "loss": 5.4378,
      "mean_token_accuracy": 0.1704568549990654,
      "num_tokens": 49873465.0,
      "step": 24035
    },
    {
      "entropy": 5.778833818435669,
      "epoch": 2.1806966618287373,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004523847267200141,
      "loss": 5.4559,
      "mean_token_accuracy": 0.1655774861574173,
      "num_tokens": 49883690.0,
      "step": 24040
    },
    {
      "entropy": 5.701267528533935,
      "epoch": 2.1811502177068216,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045236481944379165,
      "loss": 5.4223,
      "mean_token_accuracy": 0.1734446793794632,
      "num_tokens": 49894418.0,
      "step": 24045
    },
    {
      "entropy": 5.630090522766113,
      "epoch": 2.1816037735849054,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004523449084996196,
      "loss": 5.2987,
      "mean_token_accuracy": 0.18213650584220886,
      "num_tokens": 49903927.0,
      "step": 24050
    },
    {
      "entropy": 5.722528743743896,
      "epoch": 2.1820573294629897,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004523249938879096,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1722631111741066,
      "num_tokens": 49914203.0,
      "step": 24055
    },
    {
      "entropy": 5.691921377182007,
      "epoch": 2.182510885341074,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045230507560907354,
      "loss": 5.3714,
      "mean_token_accuracy": 0.17107501327991487,
      "num_tokens": 49924320.0,
      "step": 24060
    },
    {
      "entropy": 5.59493842124939,
      "epoch": 2.1829644412191582,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045228515366352334,
      "loss": 5.3021,
      "mean_token_accuracy": 0.17933637648820877,
      "num_tokens": 49934634.0,
      "step": 24065
    },
    {
      "entropy": 5.7896280765533445,
      "epoch": 2.1834179970972425,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045226522805167095,
      "loss": 5.3918,
      "mean_token_accuracy": 0.1736759752035141,
      "num_tokens": 49944586.0,
      "step": 24070
    },
    {
      "entropy": 5.72265625,
      "epoch": 2.1838715529753268,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004522452987739286,
      "loss": 5.3593,
      "mean_token_accuracy": 0.17860261499881744,
      "num_tokens": 49955441.0,
      "step": 24075
    },
    {
      "entropy": 5.777554035186768,
      "epoch": 2.1843251088534106,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045222536583070823,
      "loss": 5.4906,
      "mean_token_accuracy": 0.17508815377950668,
      "num_tokens": 49966718.0,
      "step": 24080
    },
    {
      "entropy": 5.7713369846344,
      "epoch": 2.184778664731495,
      "grad_norm": 3.78125,
      "learning_rate": 0.00045220542922242205,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16385846734046935,
      "num_tokens": 49979483.0,
      "step": 24085
    },
    {
      "entropy": 5.705483293533325,
      "epoch": 2.185232220609579,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045218548894948254,
      "loss": 5.3606,
      "mean_token_accuracy": 0.1796888530254364,
      "num_tokens": 49989976.0,
      "step": 24090
    },
    {
      "entropy": 5.695874977111816,
      "epoch": 2.1856857764876634,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045216554501230187,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17967433482408524,
      "num_tokens": 50000069.0,
      "step": 24095
    },
    {
      "entropy": 5.716794109344482,
      "epoch": 2.1861393323657476,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004521455974112927,
      "loss": 5.35,
      "mean_token_accuracy": 0.17391643524169922,
      "num_tokens": 50009970.0,
      "step": 24100
    },
    {
      "entropy": 5.662805700302124,
      "epoch": 2.1865928882438315,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045212564614686735,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1808616876602173,
      "num_tokens": 50019590.0,
      "step": 24105
    },
    {
      "entropy": 5.690664863586425,
      "epoch": 2.1870464441219157,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004521056912194386,
      "loss": 5.3722,
      "mean_token_accuracy": 0.16459959894418716,
      "num_tokens": 50030092.0,
      "step": 24110
    },
    {
      "entropy": 5.714944648742676,
      "epoch": 2.1875,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004520857326294189,
      "loss": 5.379,
      "mean_token_accuracy": 0.17063674479722976,
      "num_tokens": 50040318.0,
      "step": 24115
    },
    {
      "entropy": 5.752329969406128,
      "epoch": 2.1879535558780843,
      "grad_norm": 3.1875,
      "learning_rate": 0.00045206577037722114,
      "loss": 5.4865,
      "mean_token_accuracy": 0.160724475979805,
      "num_tokens": 50050507.0,
      "step": 24120
    },
    {
      "entropy": 5.813521194458008,
      "epoch": 2.1884071117561685,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004520458044632582,
      "loss": 5.4285,
      "mean_token_accuracy": 0.17264964580535888,
      "num_tokens": 50059727.0,
      "step": 24125
    },
    {
      "entropy": 5.737054967880249,
      "epoch": 2.1888606676342524,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045202583488794285,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17207375913858414,
      "num_tokens": 50070812.0,
      "step": 24130
    },
    {
      "entropy": 5.705451583862304,
      "epoch": 2.1893142235123366,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004520058616516881,
      "loss": 5.4831,
      "mean_token_accuracy": 0.16498000770807267,
      "num_tokens": 50081175.0,
      "step": 24135
    },
    {
      "entropy": 5.683077478408814,
      "epoch": 2.189767779390421,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045198588475490716,
      "loss": 5.3485,
      "mean_token_accuracy": 0.17224738001823425,
      "num_tokens": 50092246.0,
      "step": 24140
    },
    {
      "entropy": 5.749605369567871,
      "epoch": 2.190221335268505,
      "grad_norm": 1.5,
      "learning_rate": 0.0004519659041980129,
      "loss": 5.4215,
      "mean_token_accuracy": 0.1745454803109169,
      "num_tokens": 50101934.0,
      "step": 24145
    },
    {
      "entropy": 5.725244808197021,
      "epoch": 2.1906748911465894,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004519459199814187,
      "loss": 5.4175,
      "mean_token_accuracy": 0.16930686235427855,
      "num_tokens": 50112159.0,
      "step": 24150
    },
    {
      "entropy": 5.684139156341553,
      "epoch": 2.1911284470246732,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004519259321055377,
      "loss": 5.3949,
      "mean_token_accuracy": 0.1734326034784317,
      "num_tokens": 50123079.0,
      "step": 24155
    },
    {
      "entropy": 5.672390460968018,
      "epoch": 2.1915820029027575,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004519059405707834,
      "loss": 5.3635,
      "mean_token_accuracy": 0.17133008539676667,
      "num_tokens": 50133097.0,
      "step": 24160
    },
    {
      "entropy": 5.813031578063965,
      "epoch": 2.1920355587808418,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045188594537756916,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1730606496334076,
      "num_tokens": 50143062.0,
      "step": 24165
    },
    {
      "entropy": 5.753677177429199,
      "epoch": 2.192489114658926,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045186594652630853,
      "loss": 5.4023,
      "mean_token_accuracy": 0.1711995005607605,
      "num_tokens": 50154482.0,
      "step": 24170
    },
    {
      "entropy": 5.702457046508789,
      "epoch": 2.1929426705370103,
      "grad_norm": 2.140625,
      "learning_rate": 0.000451845944017415,
      "loss": 5.3672,
      "mean_token_accuracy": 0.17500023245811464,
      "num_tokens": 50164880.0,
      "step": 24175
    },
    {
      "entropy": 5.655025768280029,
      "epoch": 2.1933962264150946,
      "grad_norm": 1.75,
      "learning_rate": 0.00045182593785130225,
      "loss": 5.293,
      "mean_token_accuracy": 0.18271662592887877,
      "num_tokens": 50173896.0,
      "step": 24180
    },
    {
      "entropy": 5.733198976516723,
      "epoch": 2.1938497822931784,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004518059280283842,
      "loss": 5.4394,
      "mean_token_accuracy": 0.169055937230587,
      "num_tokens": 50185078.0,
      "step": 24185
    },
    {
      "entropy": 5.7075587749481205,
      "epoch": 2.1943033381712627,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045178591454907434,
      "loss": 5.4045,
      "mean_token_accuracy": 0.17482656985521317,
      "num_tokens": 50195500.0,
      "step": 24190
    },
    {
      "entropy": 5.609851503372193,
      "epoch": 2.194756894049347,
      "grad_norm": 3.46875,
      "learning_rate": 0.00045176589741378675,
      "loss": 5.3261,
      "mean_token_accuracy": 0.17330596148967742,
      "num_tokens": 50206103.0,
      "step": 24195
    },
    {
      "entropy": 5.65486512184143,
      "epoch": 2.195210449927431,
      "grad_norm": 5.5,
      "learning_rate": 0.0004517458766229354,
      "loss": 5.2904,
      "mean_token_accuracy": 0.17644621580839157,
      "num_tokens": 50216522.0,
      "step": 24200
    },
    {
      "entropy": 5.741631078720093,
      "epoch": 2.195664005805515,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045172585217693426,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17685414999723434,
      "num_tokens": 50226721.0,
      "step": 24205
    },
    {
      "entropy": 5.701233625411987,
      "epoch": 2.1961175616835993,
      "grad_norm": 2.5,
      "learning_rate": 0.00045170582407619747,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17048148810863495,
      "num_tokens": 50236327.0,
      "step": 24210
    },
    {
      "entropy": 5.656648111343384,
      "epoch": 2.1965711175616836,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004516857923211392,
      "loss": 5.3032,
      "mean_token_accuracy": 0.18677099645137787,
      "num_tokens": 50245729.0,
      "step": 24215
    },
    {
      "entropy": 5.76134295463562,
      "epoch": 2.197024673439768,
      "grad_norm": 2.34375,
      "learning_rate": 0.00045166575691217375,
      "loss": 5.4824,
      "mean_token_accuracy": 0.16424697190523146,
      "num_tokens": 50257024.0,
      "step": 24220
    },
    {
      "entropy": 5.7273242473602295,
      "epoch": 2.197478229317852,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045164571784971536,
      "loss": 5.305,
      "mean_token_accuracy": 0.17857375293970107,
      "num_tokens": 50266910.0,
      "step": 24225
    },
    {
      "entropy": 5.683621454238891,
      "epoch": 2.1979317851959363,
      "grad_norm": 4.5,
      "learning_rate": 0.00045162567513417864,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17620099782943727,
      "num_tokens": 50278233.0,
      "step": 24230
    },
    {
      "entropy": 5.640671920776367,
      "epoch": 2.19838534107402,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045160562876597776,
      "loss": 5.3466,
      "mean_token_accuracy": 0.1675179123878479,
      "num_tokens": 50289276.0,
      "step": 24235
    },
    {
      "entropy": 5.674018239974975,
      "epoch": 2.1988388969521044,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004515855787455276,
      "loss": 5.3553,
      "mean_token_accuracy": 0.17394886463880538,
      "num_tokens": 50299336.0,
      "step": 24240
    },
    {
      "entropy": 5.6761559009552,
      "epoch": 2.1992924528301887,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004515655250732426,
      "loss": 5.3081,
      "mean_token_accuracy": 0.17887204885482788,
      "num_tokens": 50308308.0,
      "step": 24245
    },
    {
      "entropy": 5.671218299865723,
      "epoch": 2.199746008708273,
      "grad_norm": 4.1875,
      "learning_rate": 0.00045154546774953764,
      "loss": 5.3477,
      "mean_token_accuracy": 0.17774360179901122,
      "num_tokens": 50318295.0,
      "step": 24250
    },
    {
      "entropy": 5.62082347869873,
      "epoch": 2.2001995645863572,
      "grad_norm": 2.0,
      "learning_rate": 0.0004515254067748274,
      "loss": 5.3613,
      "mean_token_accuracy": 0.18390874266624452,
      "num_tokens": 50329296.0,
      "step": 24255
    },
    {
      "entropy": 5.648769378662109,
      "epoch": 2.200653120464441,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004515053421495267,
      "loss": 5.4126,
      "mean_token_accuracy": 0.1731196627020836,
      "num_tokens": 50339389.0,
      "step": 24260
    },
    {
      "entropy": 5.734126567840576,
      "epoch": 2.2011066763425253,
      "grad_norm": 2.46875,
      "learning_rate": 0.00045148527387405056,
      "loss": 5.438,
      "mean_token_accuracy": 0.16901693791151046,
      "num_tokens": 50349647.0,
      "step": 24265
    },
    {
      "entropy": 5.7903149127960205,
      "epoch": 2.2015602322206096,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004514652019488141,
      "loss": 5.4484,
      "mean_token_accuracy": 0.16846761852502823,
      "num_tokens": 50361292.0,
      "step": 24270
    },
    {
      "entropy": 5.63925461769104,
      "epoch": 2.202013788098694,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004514451263742321,
      "loss": 5.3362,
      "mean_token_accuracy": 0.17345838397741317,
      "num_tokens": 50372293.0,
      "step": 24275
    },
    {
      "entropy": 5.625564432144165,
      "epoch": 2.202467343976778,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004514250471507201,
      "loss": 5.3213,
      "mean_token_accuracy": 0.1827323243021965,
      "num_tokens": 50382462.0,
      "step": 24280
    },
    {
      "entropy": 5.671874904632569,
      "epoch": 2.202920899854862,
      "grad_norm": 2.765625,
      "learning_rate": 0.00045140496427869306,
      "loss": 5.4062,
      "mean_token_accuracy": 0.18245164006948472,
      "num_tokens": 50393157.0,
      "step": 24285
    },
    {
      "entropy": 5.627283716201783,
      "epoch": 2.203374455732946,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004513848777585664,
      "loss": 5.2962,
      "mean_token_accuracy": 0.18019008934497832,
      "num_tokens": 50403474.0,
      "step": 24290
    },
    {
      "entropy": 5.709465265274048,
      "epoch": 2.2038280116110305,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004513647875907556,
      "loss": 5.3541,
      "mean_token_accuracy": 0.17353731095790864,
      "num_tokens": 50414337.0,
      "step": 24295
    },
    {
      "entropy": 5.7640307426452635,
      "epoch": 2.2042815674891147,
      "grad_norm": 2.6875,
      "learning_rate": 0.00045134469377567597,
      "loss": 5.4819,
      "mean_token_accuracy": 0.165507036447525,
      "num_tokens": 50424863.0,
      "step": 24300
    },
    {
      "entropy": 5.706216812133789,
      "epoch": 2.204735123367199,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004513245963137432,
      "loss": 5.3991,
      "mean_token_accuracy": 0.1769011914730072,
      "num_tokens": 50435982.0,
      "step": 24305
    },
    {
      "entropy": 5.681708526611328,
      "epoch": 2.205188679245283,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004513044952053728,
      "loss": 5.277,
      "mean_token_accuracy": 0.18039098083972932,
      "num_tokens": 50445112.0,
      "step": 24310
    },
    {
      "entropy": 5.613780212402344,
      "epoch": 2.205642235123367,
      "grad_norm": 3.515625,
      "learning_rate": 0.00045128439045098045,
      "loss": 5.3072,
      "mean_token_accuracy": 0.18537722080945968,
      "num_tokens": 50455266.0,
      "step": 24315
    },
    {
      "entropy": 5.705266666412354,
      "epoch": 2.2060957910014514,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045126428205098207,
      "loss": 5.4394,
      "mean_token_accuracy": 0.16762536615133286,
      "num_tokens": 50466930.0,
      "step": 24320
    },
    {
      "entropy": 5.715097427368164,
      "epoch": 2.2065493468795356,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004512441700057934,
      "loss": 5.3672,
      "mean_token_accuracy": 0.17923330515623093,
      "num_tokens": 50477534.0,
      "step": 24325
    },
    {
      "entropy": 5.688203430175781,
      "epoch": 2.20700290275762,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004512240543158305,
      "loss": 5.3444,
      "mean_token_accuracy": 0.1754288613796234,
      "num_tokens": 50487352.0,
      "step": 24330
    },
    {
      "entropy": 5.652774047851563,
      "epoch": 2.2074564586357037,
      "grad_norm": 2.96875,
      "learning_rate": 0.00045120393498150905,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18297676891088485,
      "num_tokens": 50496227.0,
      "step": 24335
    },
    {
      "entropy": 5.679080009460449,
      "epoch": 2.207910014513788,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045118381200324537,
      "loss": 5.3929,
      "mean_token_accuracy": 0.17054211348295212,
      "num_tokens": 50507536.0,
      "step": 24340
    },
    {
      "entropy": 5.686055135726929,
      "epoch": 2.2083635703918723,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004511636853814556,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17563130855560302,
      "num_tokens": 50517659.0,
      "step": 24345
    },
    {
      "entropy": 5.675984048843384,
      "epoch": 2.2088171262699565,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045114355511655595,
      "loss": 5.3552,
      "mean_token_accuracy": 0.17039137333631516,
      "num_tokens": 50528717.0,
      "step": 24350
    },
    {
      "entropy": 5.669805955886841,
      "epoch": 2.209270682148041,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045112342120896267,
      "loss": 5.2695,
      "mean_token_accuracy": 0.17909680753946305,
      "num_tokens": 50538449.0,
      "step": 24355
    },
    {
      "entropy": 5.712111949920654,
      "epoch": 2.2097242380261246,
      "grad_norm": 3.421875,
      "learning_rate": 0.00045110328365909215,
      "loss": 5.4212,
      "mean_token_accuracy": 0.17233389765024185,
      "num_tokens": 50549098.0,
      "step": 24360
    },
    {
      "entropy": 5.684254360198975,
      "epoch": 2.210177793904209,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004510831424673609,
      "loss": 5.4249,
      "mean_token_accuracy": 0.16957329213619232,
      "num_tokens": 50560047.0,
      "step": 24365
    },
    {
      "entropy": 5.672986555099487,
      "epoch": 2.210631349782293,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004510629976341853,
      "loss": 5.3331,
      "mean_token_accuracy": 0.1738471880555153,
      "num_tokens": 50569704.0,
      "step": 24370
    },
    {
      "entropy": 5.708509922027588,
      "epoch": 2.2110849056603774,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045104284915998215,
      "loss": 5.3802,
      "mean_token_accuracy": 0.17352173775434493,
      "num_tokens": 50580006.0,
      "step": 24375
    },
    {
      "entropy": 5.633798408508301,
      "epoch": 2.2115384615384617,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045102269704516793,
      "loss": 5.3434,
      "mean_token_accuracy": 0.18047112971544266,
      "num_tokens": 50589593.0,
      "step": 24380
    },
    {
      "entropy": 5.695724010467529,
      "epoch": 2.211992017416546,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004510025412901595,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17536103427410127,
      "num_tokens": 50598774.0,
      "step": 24385
    },
    {
      "entropy": 5.624537467956543,
      "epoch": 2.2124455732946298,
      "grad_norm": 4.59375,
      "learning_rate": 0.00045098238189537367,
      "loss": 5.2801,
      "mean_token_accuracy": 0.18431791961193084,
      "num_tokens": 50608815.0,
      "step": 24390
    },
    {
      "entropy": 5.558156251907349,
      "epoch": 2.212899129172714,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045096221886122733,
      "loss": 5.4629,
      "mean_token_accuracy": 0.17542963474988937,
      "num_tokens": 50619255.0,
      "step": 24395
    },
    {
      "entropy": 5.7424798011779785,
      "epoch": 2.2133526850507983,
      "grad_norm": 6.9375,
      "learning_rate": 0.0004509420521881375,
      "loss": 5.366,
      "mean_token_accuracy": 0.17668991088867186,
      "num_tokens": 50629483.0,
      "step": 24400
    },
    {
      "entropy": 5.722197008132935,
      "epoch": 2.2138062409288826,
      "grad_norm": 2.625,
      "learning_rate": 0.0004509218818765211,
      "loss": 5.2906,
      "mean_token_accuracy": 0.17781635075807573,
      "num_tokens": 50638552.0,
      "step": 24405
    },
    {
      "entropy": 5.728846168518066,
      "epoch": 2.2142597968069664,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045090170792679544,
      "loss": 5.2909,
      "mean_token_accuracy": 0.1744404837489128,
      "num_tokens": 50648752.0,
      "step": 24410
    },
    {
      "entropy": 5.689505243301392,
      "epoch": 2.2147133526850507,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045088153033937765,
      "loss": 5.3452,
      "mean_token_accuracy": 0.17550210207700728,
      "num_tokens": 50658373.0,
      "step": 24415
    },
    {
      "entropy": 5.608656167984009,
      "epoch": 2.215166908563135,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045086134911468496,
      "loss": 5.3769,
      "mean_token_accuracy": 0.18257151246070863,
      "num_tokens": 50668529.0,
      "step": 24420
    },
    {
      "entropy": 5.686266231536865,
      "epoch": 2.215620464441219,
      "grad_norm": 3.125,
      "learning_rate": 0.0004508411642531347,
      "loss": 5.3275,
      "mean_token_accuracy": 0.17822608798742295,
      "num_tokens": 50678038.0,
      "step": 24425
    },
    {
      "entropy": 5.725566482543945,
      "epoch": 2.2160740203193035,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004508209757551445,
      "loss": 5.3516,
      "mean_token_accuracy": 0.18295255899429322,
      "num_tokens": 50687788.0,
      "step": 24430
    },
    {
      "entropy": 5.691188669204712,
      "epoch": 2.2165275761973877,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045080078362113153,
      "loss": 5.3375,
      "mean_token_accuracy": 0.17383185774087906,
      "num_tokens": 50697748.0,
      "step": 24435
    },
    {
      "entropy": 5.566549253463745,
      "epoch": 2.2169811320754715,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045078058785151363,
      "loss": 5.285,
      "mean_token_accuracy": 0.18334580212831497,
      "num_tokens": 50707776.0,
      "step": 24440
    },
    {
      "entropy": 5.678296661376953,
      "epoch": 2.217434687953556,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004507603884467085,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17012028247117997,
      "num_tokens": 50717547.0,
      "step": 24445
    },
    {
      "entropy": 5.67290472984314,
      "epoch": 2.21788824383164,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004507401854071336,
      "loss": 5.2997,
      "mean_token_accuracy": 0.1805707812309265,
      "num_tokens": 50728446.0,
      "step": 24450
    },
    {
      "entropy": 5.644343662261963,
      "epoch": 2.2183417997097243,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045071997873320687,
      "loss": 5.2656,
      "mean_token_accuracy": 0.17430561184883117,
      "num_tokens": 50737909.0,
      "step": 24455
    },
    {
      "entropy": 5.631827735900879,
      "epoch": 2.2187953555878086,
      "grad_norm": 2.234375,
      "learning_rate": 0.00045069976842534635,
      "loss": 5.3826,
      "mean_token_accuracy": 0.1776716962456703,
      "num_tokens": 50749745.0,
      "step": 24460
    },
    {
      "entropy": 5.6705504894256595,
      "epoch": 2.2192489114658924,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004506795544839697,
      "loss": 5.3157,
      "mean_token_accuracy": 0.179401259124279,
      "num_tokens": 50760073.0,
      "step": 24465
    },
    {
      "entropy": 5.703800106048584,
      "epoch": 2.2197024673439767,
      "grad_norm": 1.625,
      "learning_rate": 0.0004506593369094951,
      "loss": 5.3037,
      "mean_token_accuracy": 0.1770712286233902,
      "num_tokens": 50769261.0,
      "step": 24470
    },
    {
      "entropy": 5.730454635620117,
      "epoch": 2.220156023222061,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045063911570234074,
      "loss": 5.438,
      "mean_token_accuracy": 0.1670236200094223,
      "num_tokens": 50779572.0,
      "step": 24475
    },
    {
      "entropy": 5.6452878475189205,
      "epoch": 2.2206095791001452,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004506188908629247,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17840782254934312,
      "num_tokens": 50789893.0,
      "step": 24480
    },
    {
      "entropy": 5.582854270935059,
      "epoch": 2.2210631349782295,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004505986623916652,
      "loss": 5.2391,
      "mean_token_accuracy": 0.17784329503774643,
      "num_tokens": 50799177.0,
      "step": 24485
    },
    {
      "entropy": 5.670619249343872,
      "epoch": 2.2215166908563133,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004505784302889806,
      "loss": 5.3564,
      "mean_token_accuracy": 0.18006177842617035,
      "num_tokens": 50810093.0,
      "step": 24490
    },
    {
      "entropy": 5.772902202606201,
      "epoch": 2.2219702467343976,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045055819455528935,
      "loss": 5.4897,
      "mean_token_accuracy": 0.16954036951065063,
      "num_tokens": 50820272.0,
      "step": 24495
    },
    {
      "entropy": 5.703845643997193,
      "epoch": 2.222423802612482,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004505379551910099,
      "loss": 5.4118,
      "mean_token_accuracy": 0.17841194123029708,
      "num_tokens": 50830301.0,
      "step": 24500
    },
    {
      "entropy": 5.714677333831787,
      "epoch": 2.222877358490566,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004505177121965608,
      "loss": 5.3852,
      "mean_token_accuracy": 0.17786646783351898,
      "num_tokens": 50839937.0,
      "step": 24505
    },
    {
      "entropy": 5.652593803405762,
      "epoch": 2.2233309143686504,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045049746557236067,
      "loss": 5.3156,
      "mean_token_accuracy": 0.18092899322509765,
      "num_tokens": 50850295.0,
      "step": 24510
    },
    {
      "entropy": 5.711698007583618,
      "epoch": 2.223784470246734,
      "grad_norm": 2.875,
      "learning_rate": 0.00045047721531882827,
      "loss": 5.3804,
      "mean_token_accuracy": 0.16884447634220123,
      "num_tokens": 50861889.0,
      "step": 24515
    },
    {
      "entropy": 5.693487358093262,
      "epoch": 2.2242380261248185,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004504569614363823,
      "loss": 5.3532,
      "mean_token_accuracy": 0.17613654136657714,
      "num_tokens": 50872011.0,
      "step": 24520
    },
    {
      "entropy": 5.758451509475708,
      "epoch": 2.2246915820029027,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004504367039254416,
      "loss": 5.4035,
      "mean_token_accuracy": 0.17225524187088012,
      "num_tokens": 50882559.0,
      "step": 24525
    },
    {
      "entropy": 5.6926459789276125,
      "epoch": 2.225145137880987,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045041644278642523,
      "loss": 5.435,
      "mean_token_accuracy": 0.16741650700569152,
      "num_tokens": 50894132.0,
      "step": 24530
    },
    {
      "entropy": 5.719705581665039,
      "epoch": 2.2255986937590713,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045039617801975214,
      "loss": 5.3839,
      "mean_token_accuracy": 0.174501995742321,
      "num_tokens": 50904337.0,
      "step": 24535
    },
    {
      "entropy": 5.732735013961792,
      "epoch": 2.2260522496371555,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045037590962584143,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16616521626710892,
      "num_tokens": 50914961.0,
      "step": 24540
    },
    {
      "entropy": 5.698552465438842,
      "epoch": 2.2265058055152394,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004503556376051121,
      "loss": 5.3504,
      "mean_token_accuracy": 0.1770284131169319,
      "num_tokens": 50925682.0,
      "step": 24545
    },
    {
      "entropy": 5.687218093872071,
      "epoch": 2.2269593613933236,
      "grad_norm": 1.984375,
      "learning_rate": 0.00045033536195798355,
      "loss": 5.2592,
      "mean_token_accuracy": 0.1854511618614197,
      "num_tokens": 50935809.0,
      "step": 24550
    },
    {
      "entropy": 5.792795324325562,
      "epoch": 2.227412917271408,
      "grad_norm": 3.390625,
      "learning_rate": 0.000450315082684875,
      "loss": 5.4723,
      "mean_token_accuracy": 0.16456380635499954,
      "num_tokens": 50946098.0,
      "step": 24555
    },
    {
      "entropy": 5.730971479415894,
      "epoch": 2.227866473149492,
      "grad_norm": 2.375,
      "learning_rate": 0.0004502947997862059,
      "loss": 5.4224,
      "mean_token_accuracy": 0.17428476512432098,
      "num_tokens": 50956240.0,
      "step": 24560
    },
    {
      "entropy": 5.6831506252288815,
      "epoch": 2.228320029027576,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004502745132623957,
      "loss": 5.3208,
      "mean_token_accuracy": 0.18199390470981597,
      "num_tokens": 50966870.0,
      "step": 24565
    },
    {
      "entropy": 5.6619256973266605,
      "epoch": 2.2287735849056602,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004502542231138639,
      "loss": 5.3481,
      "mean_token_accuracy": 0.1742341011762619,
      "num_tokens": 50977528.0,
      "step": 24570
    },
    {
      "entropy": 5.664371299743652,
      "epoch": 2.2292271407837445,
      "grad_norm": 2.34375,
      "learning_rate": 0.00045023392934103005,
      "loss": 5.3452,
      "mean_token_accuracy": 0.17258205711841584,
      "num_tokens": 50988565.0,
      "step": 24575
    },
    {
      "entropy": 5.684770202636718,
      "epoch": 2.2296806966618288,
      "grad_norm": 1.75,
      "learning_rate": 0.00045021363194431397,
      "loss": 5.3902,
      "mean_token_accuracy": 0.18054822236299514,
      "num_tokens": 50998410.0,
      "step": 24580
    },
    {
      "entropy": 5.716803646087646,
      "epoch": 2.230134252539913,
      "grad_norm": 1.875,
      "learning_rate": 0.0004501933309241352,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17064201533794404,
      "num_tokens": 51008890.0,
      "step": 24585
    },
    {
      "entropy": 5.65361180305481,
      "epoch": 2.2305878084179973,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004501730262809138,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17451439350843428,
      "num_tokens": 51020264.0,
      "step": 24590
    },
    {
      "entropy": 5.662713575363159,
      "epoch": 2.231041364296081,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004501527180150696,
      "loss": 5.2596,
      "mean_token_accuracy": 0.18787623643875123,
      "num_tokens": 51029892.0,
      "step": 24595
    },
    {
      "entropy": 5.738155555725098,
      "epoch": 2.2314949201741654,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004501324061270224,
      "loss": 5.4884,
      "mean_token_accuracy": 0.17241487056016921,
      "num_tokens": 51040499.0,
      "step": 24600
    },
    {
      "entropy": 5.583428049087525,
      "epoch": 2.2319484760522497,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004501120906171926,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1829586699604988,
      "num_tokens": 51050395.0,
      "step": 24605
    },
    {
      "entropy": 5.642862558364868,
      "epoch": 2.232402031930334,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045009177148600004,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1802951544523239,
      "num_tokens": 51060316.0,
      "step": 24610
    },
    {
      "entropy": 5.678933143615723,
      "epoch": 2.232855587808418,
      "grad_norm": 2.0625,
      "learning_rate": 0.00045007144873386514,
      "loss": 5.3185,
      "mean_token_accuracy": 0.18013239204883574,
      "num_tokens": 51070032.0,
      "step": 24615
    },
    {
      "entropy": 5.7910301208496096,
      "epoch": 2.233309143686502,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045005112236120795,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16794305592775344,
      "num_tokens": 51080738.0,
      "step": 24620
    },
    {
      "entropy": 5.675151109695435,
      "epoch": 2.2337626995645863,
      "grad_norm": 1.640625,
      "learning_rate": 0.000450030792368449,
      "loss": 5.2934,
      "mean_token_accuracy": 0.18334188014268876,
      "num_tokens": 51090287.0,
      "step": 24625
    },
    {
      "entropy": 5.615402603149414,
      "epoch": 2.2342162554426706,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004500104587560086,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17943671345710754,
      "num_tokens": 51101188.0,
      "step": 24630
    },
    {
      "entropy": 5.633149909973144,
      "epoch": 2.234669811320755,
      "grad_norm": 2.625,
      "learning_rate": 0.0004499901215243074,
      "loss": 5.264,
      "mean_token_accuracy": 0.17770269215106965,
      "num_tokens": 51112310.0,
      "step": 24635
    },
    {
      "entropy": 5.748424339294433,
      "epoch": 2.235123367198839,
      "grad_norm": 2.984375,
      "learning_rate": 0.00044996978067376575,
      "loss": 5.4058,
      "mean_token_accuracy": 0.174511681497097,
      "num_tokens": 51122481.0,
      "step": 24640
    },
    {
      "entropy": 5.690781736373902,
      "epoch": 2.235576923076923,
      "grad_norm": 2.25,
      "learning_rate": 0.00044994943620480453,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17524023056030275,
      "num_tokens": 51131918.0,
      "step": 24645
    },
    {
      "entropy": 5.681595373153686,
      "epoch": 2.236030478955007,
      "grad_norm": 2.25,
      "learning_rate": 0.00044992908811784446,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16830738335847856,
      "num_tokens": 51142861.0,
      "step": 24650
    },
    {
      "entropy": 5.729560995101929,
      "epoch": 2.2364840348330914,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044990873641330615,
      "loss": 5.3881,
      "mean_token_accuracy": 0.17312805503606796,
      "num_tokens": 51153956.0,
      "step": 24655
    },
    {
      "entropy": 5.691422319412231,
      "epoch": 2.2369375907111757,
      "grad_norm": 4.875,
      "learning_rate": 0.0004498883810916107,
      "loss": 5.2909,
      "mean_token_accuracy": 0.17629409879446029,
      "num_tokens": 51164499.0,
      "step": 24660
    },
    {
      "entropy": 5.666657114028931,
      "epoch": 2.23739114658926,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004498680221531789,
      "loss": 5.4236,
      "mean_token_accuracy": 0.1735670268535614,
      "num_tokens": 51175722.0,
      "step": 24665
    },
    {
      "entropy": 5.6956868171691895,
      "epoch": 2.237844702467344,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004498476595984319,
      "loss": 5.3276,
      "mean_token_accuracy": 0.17978036850690843,
      "num_tokens": 51184597.0,
      "step": 24670
    },
    {
      "entropy": 5.6261210441589355,
      "epoch": 2.238298258345428,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004498272934277906,
      "loss": 5.3671,
      "mean_token_accuracy": 0.17737506777048112,
      "num_tokens": 51194751.0,
      "step": 24675
    },
    {
      "entropy": 5.66452465057373,
      "epoch": 2.2387518142235123,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00044980692364167646,
      "loss": 5.2964,
      "mean_token_accuracy": 0.18360490649938582,
      "num_tokens": 51204184.0,
      "step": 24680
    },
    {
      "entropy": 5.635484743118286,
      "epoch": 2.2392053701015966,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004497865502405105,
      "loss": 5.4218,
      "mean_token_accuracy": 0.17889947444200516,
      "num_tokens": 51213761.0,
      "step": 24685
    },
    {
      "entropy": 5.651751327514648,
      "epoch": 2.239658925979681,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044976617322471424,
      "loss": 5.3237,
      "mean_token_accuracy": 0.1783318489789963,
      "num_tokens": 51224150.0,
      "step": 24690
    },
    {
      "entropy": 5.788090944290161,
      "epoch": 2.2401124818577647,
      "grad_norm": 1.828125,
      "learning_rate": 0.000449745792594709,
      "loss": 5.4774,
      "mean_token_accuracy": 0.1736172080039978,
      "num_tokens": 51235133.0,
      "step": 24695
    },
    {
      "entropy": 5.781828212738037,
      "epoch": 2.240566037735849,
      "grad_norm": 2.46875,
      "learning_rate": 0.00044972540835091617,
      "loss": 5.3766,
      "mean_token_accuracy": 0.17676015198230743,
      "num_tokens": 51245632.0,
      "step": 24700
    },
    {
      "entropy": 5.711320400238037,
      "epoch": 2.241019593613933,
      "grad_norm": 2.125,
      "learning_rate": 0.00044970502049375736,
      "loss": 5.4034,
      "mean_token_accuracy": 0.1770749107003212,
      "num_tokens": 51255747.0,
      "step": 24705
    },
    {
      "entropy": 5.711162090301514,
      "epoch": 2.2414731494920175,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004496846290236542,
      "loss": 5.336,
      "mean_token_accuracy": 0.17857785820960997,
      "num_tokens": 51265531.0,
      "step": 24710
    },
    {
      "entropy": 5.692275333404541,
      "epoch": 2.2419267053701017,
      "grad_norm": 2.859375,
      "learning_rate": 0.00044966423394102845,
      "loss": 5.3847,
      "mean_token_accuracy": 0.17635358273983,
      "num_tokens": 51275319.0,
      "step": 24715
    },
    {
      "entropy": 5.719060945510864,
      "epoch": 2.2423802612481856,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00044964383524630176,
      "loss": 5.3754,
      "mean_token_accuracy": 0.18314924091100693,
      "num_tokens": 51285655.0,
      "step": 24720
    },
    {
      "entropy": 5.62483057975769,
      "epoch": 2.24283381712627,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004496234329398961,
      "loss": 5.2679,
      "mean_token_accuracy": 0.1768614575266838,
      "num_tokens": 51295938.0,
      "step": 24725
    },
    {
      "entropy": 5.716745805740357,
      "epoch": 2.243287373004354,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004496030270222333,
      "loss": 5.3066,
      "mean_token_accuracy": 0.17818249017000198,
      "num_tokens": 51306483.0,
      "step": 24730
    },
    {
      "entropy": 5.700893497467041,
      "epoch": 2.2437409288824384,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044958261749373547,
      "loss": 5.3478,
      "mean_token_accuracy": 0.17956360876560212,
      "num_tokens": 51316446.0,
      "step": 24735
    },
    {
      "entropy": 5.617989492416382,
      "epoch": 2.2441944847605226,
      "grad_norm": 2.546875,
      "learning_rate": 0.00044956220435482465,
      "loss": 5.3602,
      "mean_token_accuracy": 0.17592585533857347,
      "num_tokens": 51326840.0,
      "step": 24740
    },
    {
      "entropy": 5.699657773971557,
      "epoch": 2.244648040638607,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044954178760592283,
      "loss": 5.4105,
      "mean_token_accuracy": 0.16853396892547606,
      "num_tokens": 51337573.0,
      "step": 24745
    },
    {
      "entropy": 5.757193994522095,
      "epoch": 2.2451015965166907,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044952136724745254,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1742326706647873,
      "num_tokens": 51347827.0,
      "step": 24750
    },
    {
      "entropy": 5.6817864894866945,
      "epoch": 2.245555152394775,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004495009432798358,
      "loss": 5.3215,
      "mean_token_accuracy": 0.18098485171794892,
      "num_tokens": 51358396.0,
      "step": 24755
    },
    {
      "entropy": 5.76072850227356,
      "epoch": 2.2460087082728593,
      "grad_norm": 2.265625,
      "learning_rate": 0.00044948051570349503,
      "loss": 5.3533,
      "mean_token_accuracy": 0.17031417042016983,
      "num_tokens": 51368882.0,
      "step": 24760
    },
    {
      "entropy": 5.700961399078369,
      "epoch": 2.2464622641509435,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004494600845188528,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17782808393239974,
      "num_tokens": 51378717.0,
      "step": 24765
    },
    {
      "entropy": 5.651447105407715,
      "epoch": 2.2469158200290273,
      "grad_norm": 2.53125,
      "learning_rate": 0.00044943964972633154,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17191415876150132,
      "num_tokens": 51389270.0,
      "step": 24770
    },
    {
      "entropy": 5.708307218551636,
      "epoch": 2.2473693759071116,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004494192113263539,
      "loss": 5.3691,
      "mean_token_accuracy": 0.1802399531006813,
      "num_tokens": 51400025.0,
      "step": 24775
    },
    {
      "entropy": 5.718846464157105,
      "epoch": 2.247822931785196,
      "grad_norm": 2.28125,
      "learning_rate": 0.00044939876931934255,
      "loss": 5.3601,
      "mean_token_accuracy": 0.18164087533950807,
      "num_tokens": 51410903.0,
      "step": 24780
    },
    {
      "entropy": 5.597605991363525,
      "epoch": 2.24827648766328,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004493783237057201,
      "loss": 5.3162,
      "mean_token_accuracy": 0.17948150932788848,
      "num_tokens": 51420246.0,
      "step": 24785
    },
    {
      "entropy": 5.733811664581299,
      "epoch": 2.2487300435413644,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004493578744859095,
      "loss": 5.4556,
      "mean_token_accuracy": 0.17657198756933212,
      "num_tokens": 51431309.0,
      "step": 24790
    },
    {
      "entropy": 5.808437061309815,
      "epoch": 2.2491835994194487,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004493374216603338,
      "loss": 5.4405,
      "mean_token_accuracy": 0.17044740915298462,
      "num_tokens": 51443142.0,
      "step": 24795
    },
    {
      "entropy": 5.8512650489807125,
      "epoch": 2.2496371552975325,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004493169652294156,
      "loss": 5.5798,
      "mean_token_accuracy": 0.16584542840719224,
      "num_tokens": 51454499.0,
      "step": 24800
    },
    {
      "entropy": 5.710652017593384,
      "epoch": 2.2500907111756168,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044929650519357823,
      "loss": 5.3703,
      "mean_token_accuracy": 0.17560516446828842,
      "num_tokens": 51464667.0,
      "step": 24805
    },
    {
      "entropy": 5.757180309295654,
      "epoch": 2.250544267053701,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004492760415532447,
      "loss": 5.3849,
      "mean_token_accuracy": 0.1774010553956032,
      "num_tokens": 51473627.0,
      "step": 24810
    },
    {
      "entropy": 5.605583810806275,
      "epoch": 2.2509978229317853,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004492555743088382,
      "loss": 5.2327,
      "mean_token_accuracy": 0.17905665636062623,
      "num_tokens": 51483052.0,
      "step": 24815
    },
    {
      "entropy": 5.631391954421997,
      "epoch": 2.2514513788098696,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000449235103460782,
      "loss": 5.3467,
      "mean_token_accuracy": 0.1772044852375984,
      "num_tokens": 51492422.0,
      "step": 24820
    },
    {
      "entropy": 5.634738492965698,
      "epoch": 2.2519049346879534,
      "grad_norm": 2.25,
      "learning_rate": 0.0004492146290094995,
      "loss": 5.3146,
      "mean_token_accuracy": 0.17650102823972702,
      "num_tokens": 51503566.0,
      "step": 24825
    },
    {
      "entropy": 5.647380447387695,
      "epoch": 2.2523584905660377,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044919415095541406,
      "loss": 5.3375,
      "mean_token_accuracy": 0.17699647545814515,
      "num_tokens": 51513699.0,
      "step": 24830
    },
    {
      "entropy": 5.73447995185852,
      "epoch": 2.252812046444122,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044917366929894914,
      "loss": 5.359,
      "mean_token_accuracy": 0.1696239322423935,
      "num_tokens": 51524745.0,
      "step": 24835
    },
    {
      "entropy": 5.68453574180603,
      "epoch": 2.253265602322206,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044915318404052846,
      "loss": 5.2867,
      "mean_token_accuracy": 0.17995968461036682,
      "num_tokens": 51536516.0,
      "step": 24840
    },
    {
      "entropy": 5.681751251220703,
      "epoch": 2.2537191582002905,
      "grad_norm": 2.765625,
      "learning_rate": 0.00044913269518057547,
      "loss": 5.3925,
      "mean_token_accuracy": 0.17168912291526794,
      "num_tokens": 51547063.0,
      "step": 24845
    },
    {
      "entropy": 5.667189455032348,
      "epoch": 2.2541727140783743,
      "grad_norm": 2.21875,
      "learning_rate": 0.000449112202719514,
      "loss": 5.3496,
      "mean_token_accuracy": 0.1736954003572464,
      "num_tokens": 51557244.0,
      "step": 24850
    },
    {
      "entropy": 5.729010677337646,
      "epoch": 2.2546262699564585,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004490917066577678,
      "loss": 5.4011,
      "mean_token_accuracy": 0.17844539880752563,
      "num_tokens": 51567630.0,
      "step": 24855
    },
    {
      "entropy": 5.712599611282348,
      "epoch": 2.255079825834543,
      "grad_norm": 2.390625,
      "learning_rate": 0.00044907120699576075,
      "loss": 5.4071,
      "mean_token_accuracy": 0.17598091214895248,
      "num_tokens": 51577951.0,
      "step": 24860
    },
    {
      "entropy": 5.708667993545532,
      "epoch": 2.255533381712627,
      "grad_norm": 2.671875,
      "learning_rate": 0.00044905070373391684,
      "loss": 5.3235,
      "mean_token_accuracy": 0.17908613979816437,
      "num_tokens": 51587533.0,
      "step": 24865
    },
    {
      "entropy": 5.7360148429870605,
      "epoch": 2.2559869375907113,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004490301968726599,
      "loss": 5.3207,
      "mean_token_accuracy": 0.18527463525533677,
      "num_tokens": 51597212.0,
      "step": 24870
    },
    {
      "entropy": 5.699779319763183,
      "epoch": 2.256440493468795,
      "grad_norm": 2.421875,
      "learning_rate": 0.00044900968641241426,
      "loss": 5.4017,
      "mean_token_accuracy": 0.17880237102508545,
      "num_tokens": 51608520.0,
      "step": 24875
    },
    {
      "entropy": 5.738628673553467,
      "epoch": 2.2568940493468794,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004489891723536039,
      "loss": 5.4285,
      "mean_token_accuracy": 0.17145243734121324,
      "num_tokens": 51619651.0,
      "step": 24880
    },
    {
      "entropy": 5.64633412361145,
      "epoch": 2.2573476052249637,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004489686546966531,
      "loss": 5.2651,
      "mean_token_accuracy": 0.17490866184234619,
      "num_tokens": 51630184.0,
      "step": 24885
    },
    {
      "entropy": 5.622054576873779,
      "epoch": 2.257801161103048,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044894813344198626,
      "loss": 5.2857,
      "mean_token_accuracy": 0.1829395353794098,
      "num_tokens": 51641293.0,
      "step": 24890
    },
    {
      "entropy": 5.699812173843384,
      "epoch": 2.2582547169811322,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004489276085900276,
      "loss": 5.3438,
      "mean_token_accuracy": 0.17903086096048354,
      "num_tokens": 51652002.0,
      "step": 24895
    },
    {
      "entropy": 5.578924942016601,
      "epoch": 2.2587082728592165,
      "grad_norm": 2.484375,
      "learning_rate": 0.00044890708014120174,
      "loss": 5.2257,
      "mean_token_accuracy": 0.18732941448688506,
      "num_tokens": 51661668.0,
      "step": 24900
    },
    {
      "entropy": 5.745677185058594,
      "epoch": 2.2591618287373003,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004488865480959331,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1716599464416504,
      "num_tokens": 51672879.0,
      "step": 24905
    },
    {
      "entropy": 5.693577432632447,
      "epoch": 2.2596153846153846,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004488660124546464,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17325607687234879,
      "num_tokens": 51683878.0,
      "step": 24910
    },
    {
      "entropy": 5.635989665985107,
      "epoch": 2.260068940493469,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004488454732177662,
      "loss": 5.3564,
      "mean_token_accuracy": 0.1761213019490242,
      "num_tokens": 51693971.0,
      "step": 24915
    },
    {
      "entropy": 5.6987988471984865,
      "epoch": 2.260522496371553,
      "grad_norm": 2.578125,
      "learning_rate": 0.00044882493038571737,
      "loss": 5.2533,
      "mean_token_accuracy": 0.18701134920120238,
      "num_tokens": 51703685.0,
      "step": 24920
    },
    {
      "entropy": 5.690208482742309,
      "epoch": 2.260976052249637,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044880438395892457,
      "loss": 5.3853,
      "mean_token_accuracy": 0.17440817803144454,
      "num_tokens": 51714469.0,
      "step": 24925
    },
    {
      "entropy": 5.6711749076843265,
      "epoch": 2.261429608127721,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004487838339378129,
      "loss": 5.3994,
      "mean_token_accuracy": 0.17075148820877076,
      "num_tokens": 51725283.0,
      "step": 24930
    },
    {
      "entropy": 5.7103804588317875,
      "epoch": 2.2618831640058055,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004487632803228072,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17912339717149733,
      "num_tokens": 51735628.0,
      "step": 24935
    },
    {
      "entropy": 5.578019857406616,
      "epoch": 2.2623367198838897,
      "grad_norm": 2.671875,
      "learning_rate": 0.00044874272311433265,
      "loss": 5.2279,
      "mean_token_accuracy": 0.18445414900779725,
      "num_tokens": 51745991.0,
      "step": 24940
    },
    {
      "entropy": 5.64164662361145,
      "epoch": 2.262790275761974,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004487221623128143,
      "loss": 5.28,
      "mean_token_accuracy": 0.18228453248739243,
      "num_tokens": 51756306.0,
      "step": 24945
    },
    {
      "entropy": 5.666023683547974,
      "epoch": 2.2632438316400583,
      "grad_norm": 4.40625,
      "learning_rate": 0.00044870159791867743,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1829700216650963,
      "num_tokens": 51765466.0,
      "step": 24950
    },
    {
      "entropy": 5.665311527252197,
      "epoch": 2.263697387518142,
      "grad_norm": 2.8125,
      "learning_rate": 0.00044868102993234716,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1719285488128662,
      "num_tokens": 51775526.0,
      "step": 24955
    },
    {
      "entropy": 5.69591269493103,
      "epoch": 2.2641509433962264,
      "grad_norm": 1.890625,
      "learning_rate": 0.00044866045835424896,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17417544722557068,
      "num_tokens": 51785434.0,
      "step": 24960
    },
    {
      "entropy": 5.726017618179322,
      "epoch": 2.2646044992743106,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004486398831848082,
      "loss": 5.3931,
      "mean_token_accuracy": 0.1696265459060669,
      "num_tokens": 51796047.0,
      "step": 24965
    },
    {
      "entropy": 5.693665885925293,
      "epoch": 2.265058055152395,
      "grad_norm": 2.578125,
      "learning_rate": 0.00044861930442445037,
      "loss": 5.374,
      "mean_token_accuracy": 0.17118641436100007,
      "num_tokens": 51806173.0,
      "step": 24970
    },
    {
      "entropy": 5.746839141845703,
      "epoch": 2.2655116110304787,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004485987220736012,
      "loss": 5.5294,
      "mean_token_accuracy": 0.16784699112176896,
      "num_tokens": 51817644.0,
      "step": 24975
    },
    {
      "entropy": 5.734383821487427,
      "epoch": 2.265965166908563,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004485781361326861,
      "loss": 5.4709,
      "mean_token_accuracy": 0.1704474687576294,
      "num_tokens": 51829746.0,
      "step": 24980
    },
    {
      "entropy": 5.69964861869812,
      "epoch": 2.2664187227866472,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000448557546602131,
      "loss": 5.3986,
      "mean_token_accuracy": 0.17386804819107055,
      "num_tokens": 51840868.0,
      "step": 24985
    },
    {
      "entropy": 5.723479175567627,
      "epoch": 2.2668722786647315,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044853695348236156,
      "loss": 5.2678,
      "mean_token_accuracy": 0.180449415743351,
      "num_tokens": 51851385.0,
      "step": 24990
    },
    {
      "entropy": 5.673448371887207,
      "epoch": 2.2673258345428158,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044851635677380374,
      "loss": 5.301,
      "mean_token_accuracy": 0.1733400359749794,
      "num_tokens": 51861021.0,
      "step": 24995
    },
    {
      "entropy": 5.6978747844696045,
      "epoch": 2.2677793904209,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004484957564768834,
      "loss": 5.4153,
      "mean_token_accuracy": 0.16951217800378798,
      "num_tokens": 51870688.0,
      "step": 25000
    },
    {
      "entropy": 5.644614267349243,
      "epoch": 2.268232946298984,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044847515259202657,
      "loss": 5.3358,
      "mean_token_accuracy": 0.18143062442541122,
      "num_tokens": 51880439.0,
      "step": 25005
    },
    {
      "entropy": 5.659949588775635,
      "epoch": 2.268686502177068,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00044845454511965947,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17395371049642563,
      "num_tokens": 51890861.0,
      "step": 25010
    },
    {
      "entropy": 5.7381120204925535,
      "epoch": 2.2691400580551524,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044843393406020823,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17939993143081664,
      "num_tokens": 51900687.0,
      "step": 25015
    },
    {
      "entropy": 5.735888195037842,
      "epoch": 2.2695936139332367,
      "grad_norm": 4.03125,
      "learning_rate": 0.00044841331941409884,
      "loss": 5.4898,
      "mean_token_accuracy": 0.1690266951918602,
      "num_tokens": 51911805.0,
      "step": 25020
    },
    {
      "entropy": 5.702571296691895,
      "epoch": 2.270047169811321,
      "grad_norm": 2.828125,
      "learning_rate": 0.00044839270118175793,
      "loss": 5.3024,
      "mean_token_accuracy": 0.18513621985912324,
      "num_tokens": 51922520.0,
      "step": 25025
    },
    {
      "entropy": 5.743095541000367,
      "epoch": 2.2705007256894048,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004483720793636118,
      "loss": 5.4433,
      "mean_token_accuracy": 0.16756259351968766,
      "num_tokens": 51934056.0,
      "step": 25030
    },
    {
      "entropy": 5.728262233734131,
      "epoch": 2.270954281567489,
      "grad_norm": 2.875,
      "learning_rate": 0.00044835145396008685,
      "loss": 5.4182,
      "mean_token_accuracy": 0.17400194704532623,
      "num_tokens": 51944696.0,
      "step": 25035
    },
    {
      "entropy": 5.766645717620849,
      "epoch": 2.2714078374455733,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004483308249716097,
      "loss": 5.4022,
      "mean_token_accuracy": 0.1737814113497734,
      "num_tokens": 51955111.0,
      "step": 25040
    },
    {
      "entropy": 5.652970981597901,
      "epoch": 2.2718613933236576,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004483101923986068,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17830903381109237,
      "num_tokens": 51966036.0,
      "step": 25045
    },
    {
      "entropy": 5.644937753677368,
      "epoch": 2.272314949201742,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044828955624150505,
      "loss": 5.3562,
      "mean_token_accuracy": 0.17926433682441711,
      "num_tokens": 51976060.0,
      "step": 25050
    },
    {
      "entropy": 5.693533611297608,
      "epoch": 2.2727685050798256,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004482689165007311,
      "loss": 5.3426,
      "mean_token_accuracy": 0.17515721917152405,
      "num_tokens": 51986016.0,
      "step": 25055
    },
    {
      "entropy": 5.662090635299682,
      "epoch": 2.27322206095791,
      "grad_norm": 2.203125,
      "learning_rate": 0.00044824827317671177,
      "loss": 5.289,
      "mean_token_accuracy": 0.18831997066736222,
      "num_tokens": 51994794.0,
      "step": 25060
    },
    {
      "entropy": 5.592199373245239,
      "epoch": 2.273675616835994,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000448227626269874,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1817483514547348,
      "num_tokens": 52004976.0,
      "step": 25065
    },
    {
      "entropy": 5.706680059432983,
      "epoch": 2.2741291727140784,
      "grad_norm": 3.203125,
      "learning_rate": 0.00044820697578064485,
      "loss": 5.3869,
      "mean_token_accuracy": 0.1677025780081749,
      "num_tokens": 52016380.0,
      "step": 25070
    },
    {
      "entropy": 5.74155330657959,
      "epoch": 2.2745827285921627,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004481863217094513,
      "loss": 5.3279,
      "mean_token_accuracy": 0.17477558106184005,
      "num_tokens": 52026439.0,
      "step": 25075
    },
    {
      "entropy": 5.650634384155273,
      "epoch": 2.2750362844702465,
      "grad_norm": 2.0,
      "learning_rate": 0.0004481656640567204,
      "loss": 5.2921,
      "mean_token_accuracy": 0.18007634878158568,
      "num_tokens": 52037111.0,
      "step": 25080
    },
    {
      "entropy": 5.647220659255981,
      "epoch": 2.275489840348331,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004481450028228795,
      "loss": 5.3558,
      "mean_token_accuracy": 0.174395415186882,
      "num_tokens": 52047463.0,
      "step": 25085
    },
    {
      "entropy": 5.725974416732788,
      "epoch": 2.275943396226415,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004481243380083558,
      "loss": 5.3793,
      "mean_token_accuracy": 0.17489156872034073,
      "num_tokens": 52058081.0,
      "step": 25090
    },
    {
      "entropy": 5.6333190441131595,
      "epoch": 2.2763969521044993,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004481036696135767,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17811353504657745,
      "num_tokens": 52069367.0,
      "step": 25095
    },
    {
      "entropy": 5.6955023288726805,
      "epoch": 2.2768505079825836,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004480829976389695,
      "loss": 5.3127,
      "mean_token_accuracy": 0.18054232001304626,
      "num_tokens": 52080023.0,
      "step": 25100
    },
    {
      "entropy": 5.624083948135376,
      "epoch": 2.277304063860668,
      "grad_norm": 2.296875,
      "learning_rate": 0.00044806232208496194,
      "loss": 5.3277,
      "mean_token_accuracy": 0.18179234117269516,
      "num_tokens": 52089184.0,
      "step": 25105
    },
    {
      "entropy": 5.711986684799195,
      "epoch": 2.2777576197387517,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004480416429519814,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1741299197077751,
      "num_tokens": 52098711.0,
      "step": 25110
    },
    {
      "entropy": 5.764838075637817,
      "epoch": 2.278211175616836,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004480209602404556,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1816176488995552,
      "num_tokens": 52108826.0,
      "step": 25115
    },
    {
      "entropy": 5.658444738388061,
      "epoch": 2.27866473149492,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004480002739508122,
      "loss": 5.3007,
      "mean_token_accuracy": 0.18153747022151948,
      "num_tokens": 52118934.0,
      "step": 25120
    },
    {
      "entropy": 5.600614404678344,
      "epoch": 2.2791182873730045,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004479795840834791,
      "loss": 5.3177,
      "mean_token_accuracy": 0.18511003404855728,
      "num_tokens": 52129803.0,
      "step": 25125
    },
    {
      "entropy": 5.580665016174317,
      "epoch": 2.2795718432510883,
      "grad_norm": 2.140625,
      "learning_rate": 0.00044795889063888413,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18377198427915573,
      "num_tokens": 52140070.0,
      "step": 25130
    },
    {
      "entropy": 5.766581583023071,
      "epoch": 2.2800253991291726,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004479381936174552,
      "loss": 5.5448,
      "mean_token_accuracy": 0.1644115537405014,
      "num_tokens": 52150462.0,
      "step": 25135
    },
    {
      "entropy": 5.746579647064209,
      "epoch": 2.280478955007257,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004479174930196204,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17788543552160263,
      "num_tokens": 52160552.0,
      "step": 25140
    },
    {
      "entropy": 5.577821445465088,
      "epoch": 2.280932510885341,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044789678884580767,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18601563423871995,
      "num_tokens": 52169780.0,
      "step": 25145
    },
    {
      "entropy": 5.696633768081665,
      "epoch": 2.2813860667634254,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044787608109644536,
      "loss": 5.3763,
      "mean_token_accuracy": 0.18328474313020707,
      "num_tokens": 52180055.0,
      "step": 25150
    },
    {
      "entropy": 5.684626150131225,
      "epoch": 2.2818396226415096,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004478553697719616,
      "loss": 5.3169,
      "mean_token_accuracy": 0.1842746138572693,
      "num_tokens": 52189772.0,
      "step": 25155
    },
    {
      "entropy": 5.670305013656616,
      "epoch": 2.2822931785195935,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004478346548727848,
      "loss": 5.2996,
      "mean_token_accuracy": 0.18853309899568557,
      "num_tokens": 52199608.0,
      "step": 25160
    },
    {
      "entropy": 5.641065216064453,
      "epoch": 2.2827467343976777,
      "grad_norm": 1.625,
      "learning_rate": 0.0004478139363993433,
      "loss": 5.3342,
      "mean_token_accuracy": 0.17499771863222122,
      "num_tokens": 52208236.0,
      "step": 25165
    },
    {
      "entropy": 5.704846572875977,
      "epoch": 2.283200290275762,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004477932143520655,
      "loss": 5.3669,
      "mean_token_accuracy": 0.1797839730978012,
      "num_tokens": 52218403.0,
      "step": 25170
    },
    {
      "entropy": 5.684479999542236,
      "epoch": 2.2836538461538463,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044777248873138005,
      "loss": 5.3329,
      "mean_token_accuracy": 0.17920052707195283,
      "num_tokens": 52228255.0,
      "step": 25175
    },
    {
      "entropy": 5.690886831283569,
      "epoch": 2.2841074020319305,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044775175953771547,
      "loss": 5.3692,
      "mean_token_accuracy": 0.17283204346895217,
      "num_tokens": 52237732.0,
      "step": 25180
    },
    {
      "entropy": 5.777204990386963,
      "epoch": 2.2845609579100143,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004477310267715005,
      "loss": 5.5272,
      "mean_token_accuracy": 0.16520147174596786,
      "num_tokens": 52249226.0,
      "step": 25185
    },
    {
      "entropy": 5.732043313980102,
      "epoch": 2.2850145137880986,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004477102904331639,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17888532429933549,
      "num_tokens": 52259550.0,
      "step": 25190
    },
    {
      "entropy": 5.673824214935303,
      "epoch": 2.285468069666183,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044768955052313443,
      "loss": 5.3987,
      "mean_token_accuracy": 0.17428454458713533,
      "num_tokens": 52269670.0,
      "step": 25195
    },
    {
      "entropy": 5.741723966598511,
      "epoch": 2.285921625544267,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00044766880704184104,
      "loss": 5.3622,
      "mean_token_accuracy": 0.180657759308815,
      "num_tokens": 52280135.0,
      "step": 25200
    },
    {
      "entropy": 5.668017721176147,
      "epoch": 2.2863751814223514,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004476480599897128,
      "loss": 5.3015,
      "mean_token_accuracy": 0.18195862621068953,
      "num_tokens": 52291806.0,
      "step": 25205
    },
    {
      "entropy": 5.578812789916992,
      "epoch": 2.2868287373004352,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004476273093671786,
      "loss": 5.3053,
      "mean_token_accuracy": 0.17826635241508484,
      "num_tokens": 52302118.0,
      "step": 25210
    },
    {
      "entropy": 5.761087894439697,
      "epoch": 2.2872822931785195,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044760655517466766,
      "loss": 5.4316,
      "mean_token_accuracy": 0.1667913407087326,
      "num_tokens": 52312762.0,
      "step": 25215
    },
    {
      "entropy": 5.73091869354248,
      "epoch": 2.2877358490566038,
      "grad_norm": 2.125,
      "learning_rate": 0.00044758579741260916,
      "loss": 5.3969,
      "mean_token_accuracy": 0.17055546194314958,
      "num_tokens": 52323367.0,
      "step": 25220
    },
    {
      "entropy": 5.741938352584839,
      "epoch": 2.288189404934688,
      "grad_norm": 3.21875,
      "learning_rate": 0.00044756503608143247,
      "loss": 5.4272,
      "mean_token_accuracy": 0.16824697256088256,
      "num_tokens": 52335231.0,
      "step": 25225
    },
    {
      "entropy": 5.753727197647095,
      "epoch": 2.2886429608127723,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044754427118156675,
      "loss": 5.3202,
      "mean_token_accuracy": 0.17838436514139175,
      "num_tokens": 52345066.0,
      "step": 25230
    },
    {
      "entropy": 5.706684684753418,
      "epoch": 2.289096516690856,
      "grad_norm": 2.0,
      "learning_rate": 0.0004475235027134416,
      "loss": 5.3645,
      "mean_token_accuracy": 0.17343364655971527,
      "num_tokens": 52356383.0,
      "step": 25235
    },
    {
      "entropy": 5.682183313369751,
      "epoch": 2.2895500725689404,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004475027306774865,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1838977813720703,
      "num_tokens": 52366013.0,
      "step": 25240
    },
    {
      "entropy": 5.685575532913208,
      "epoch": 2.2900036284470247,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044748195507413095,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17868764698505402,
      "num_tokens": 52376298.0,
      "step": 25245
    },
    {
      "entropy": 5.591705560684204,
      "epoch": 2.290457184325109,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044746117590380456,
      "loss": 5.2719,
      "mean_token_accuracy": 0.1773749753832817,
      "num_tokens": 52387134.0,
      "step": 25250
    },
    {
      "entropy": 5.656711053848267,
      "epoch": 2.290910740203193,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004474403931669371,
      "loss": 5.2548,
      "mean_token_accuracy": 0.17900755554437636,
      "num_tokens": 52397389.0,
      "step": 25255
    },
    {
      "entropy": 5.672642993927002,
      "epoch": 2.2913642960812775,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004474196068639584,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18092149794101714,
      "num_tokens": 52407456.0,
      "step": 25260
    },
    {
      "entropy": 5.658842086791992,
      "epoch": 2.2918178519593613,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004473988169952983,
      "loss": 5.3433,
      "mean_token_accuracy": 0.17904432862997055,
      "num_tokens": 52417967.0,
      "step": 25265
    },
    {
      "entropy": 5.785518503189087,
      "epoch": 2.2922714078374455,
      "grad_norm": 1.75,
      "learning_rate": 0.00044737802356138674,
      "loss": 5.4601,
      "mean_token_accuracy": 0.17757325619459152,
      "num_tokens": 52428757.0,
      "step": 25270
    },
    {
      "entropy": 5.731307601928711,
      "epoch": 2.29272496371553,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004473572265626537,
      "loss": 5.3549,
      "mean_token_accuracy": 0.18133081793785094,
      "num_tokens": 52438973.0,
      "step": 25275
    },
    {
      "entropy": 5.611259651184082,
      "epoch": 2.293178519593614,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004473364259995293,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17793068438768386,
      "num_tokens": 52449323.0,
      "step": 25280
    },
    {
      "entropy": 5.680596113204956,
      "epoch": 2.293632075471698,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004473156218724436,
      "loss": 5.3466,
      "mean_token_accuracy": 0.18513886928558348,
      "num_tokens": 52459299.0,
      "step": 25285
    },
    {
      "entropy": 5.647358465194702,
      "epoch": 2.294085631349782,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044729481418182704,
      "loss": 5.3099,
      "mean_token_accuracy": 0.17194937467575072,
      "num_tokens": 52470026.0,
      "step": 25290
    },
    {
      "entropy": 5.705062580108643,
      "epoch": 2.2945391872278664,
      "grad_norm": 2.3125,
      "learning_rate": 0.00044727400292810985,
      "loss": 5.3004,
      "mean_token_accuracy": 0.18091041296720506,
      "num_tokens": 52479975.0,
      "step": 25295
    },
    {
      "entropy": 5.730549669265747,
      "epoch": 2.2949927431059507,
      "grad_norm": 2.421875,
      "learning_rate": 0.00044725318811172227,
      "loss": 5.4546,
      "mean_token_accuracy": 0.1740449994802475,
      "num_tokens": 52491067.0,
      "step": 25300
    },
    {
      "entropy": 5.60552945137024,
      "epoch": 2.295446298984035,
      "grad_norm": 1.71875,
      "learning_rate": 0.000447232369733095,
      "loss": 5.3094,
      "mean_token_accuracy": 0.18256543576717377,
      "num_tokens": 52500671.0,
      "step": 25305
    },
    {
      "entropy": 5.677025985717774,
      "epoch": 2.2958998548621192,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004472115477926583,
      "loss": 5.282,
      "mean_token_accuracy": 0.17871612310409546,
      "num_tokens": 52509910.0,
      "step": 25310
    },
    {
      "entropy": 5.698044776916504,
      "epoch": 2.296353410740203,
      "grad_norm": 6.125,
      "learning_rate": 0.00044719072229084295,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17954202145338058,
      "num_tokens": 52520517.0,
      "step": 25315
    },
    {
      "entropy": 5.701544904708863,
      "epoch": 2.2968069666182873,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004471698932280796,
      "loss": 5.4075,
      "mean_token_accuracy": 0.18151070177555084,
      "num_tokens": 52530760.0,
      "step": 25320
    },
    {
      "entropy": 5.764336442947387,
      "epoch": 2.2972605224963716,
      "grad_norm": 3.046875,
      "learning_rate": 0.000447149060604799,
      "loss": 5.4306,
      "mean_token_accuracy": 0.17163496762514113,
      "num_tokens": 52541791.0,
      "step": 25325
    },
    {
      "entropy": 5.7742949485778805,
      "epoch": 2.297714078374456,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044712822442143194,
      "loss": 5.4189,
      "mean_token_accuracy": 0.1782807558774948,
      "num_tokens": 52551885.0,
      "step": 25330
    },
    {
      "entropy": 5.578240156173706,
      "epoch": 2.2981676342525397,
      "grad_norm": 2.046875,
      "learning_rate": 0.00044710738467840927,
      "loss": 5.3248,
      "mean_token_accuracy": 0.17883899956941604,
      "num_tokens": 52563080.0,
      "step": 25335
    },
    {
      "entropy": 5.663577127456665,
      "epoch": 2.298621190130624,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004470865413761621,
      "loss": 5.2811,
      "mean_token_accuracy": 0.1839486300945282,
      "num_tokens": 52573564.0,
      "step": 25340
    },
    {
      "entropy": 5.694223165512085,
      "epoch": 2.299074746008708,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044706569451512143,
      "loss": 5.341,
      "mean_token_accuracy": 0.17682420909404756,
      "num_tokens": 52583801.0,
      "step": 25345
    },
    {
      "entropy": 5.716454124450683,
      "epoch": 2.2995283018867925,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044704484409571825,
      "loss": 5.3792,
      "mean_token_accuracy": 0.17615263313055038,
      "num_tokens": 52594908.0,
      "step": 25350
    },
    {
      "entropy": 5.61153621673584,
      "epoch": 2.2999818577648767,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044702399011838393,
      "loss": 5.2785,
      "mean_token_accuracy": 0.17446652352809905,
      "num_tokens": 52604861.0,
      "step": 25355
    },
    {
      "entropy": 5.650022268295288,
      "epoch": 2.300435413642961,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004470031325835497,
      "loss": 5.354,
      "mean_token_accuracy": 0.17902897149324418,
      "num_tokens": 52616203.0,
      "step": 25360
    },
    {
      "entropy": 5.667253684997559,
      "epoch": 2.300888969521045,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004469822714916468,
      "loss": 5.2773,
      "mean_token_accuracy": 0.18675442039966583,
      "num_tokens": 52625838.0,
      "step": 25365
    },
    {
      "entropy": 5.5983106136322025,
      "epoch": 2.301342525399129,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004469614068431067,
      "loss": 5.307,
      "mean_token_accuracy": 0.18133770078420638,
      "num_tokens": 52635741.0,
      "step": 25370
    },
    {
      "entropy": 5.617379379272461,
      "epoch": 2.3017960812772134,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004469405386383608,
      "loss": 5.288,
      "mean_token_accuracy": 0.17908272743225098,
      "num_tokens": 52645948.0,
      "step": 25375
    },
    {
      "entropy": 5.663857316970825,
      "epoch": 2.3022496371552976,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044691966687784087,
      "loss": 5.3304,
      "mean_token_accuracy": 0.18033190369606017,
      "num_tokens": 52655073.0,
      "step": 25380
    },
    {
      "entropy": 5.639379072189331,
      "epoch": 2.302703193033382,
      "grad_norm": 2.25,
      "learning_rate": 0.00044689879156197833,
      "loss": 5.3374,
      "mean_token_accuracy": 0.18447690308094025,
      "num_tokens": 52664429.0,
      "step": 25385
    },
    {
      "entropy": 5.789876270294189,
      "epoch": 2.3031567489114657,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044687791269120504,
      "loss": 5.416,
      "mean_token_accuracy": 0.1737708181142807,
      "num_tokens": 52674890.0,
      "step": 25390
    },
    {
      "entropy": 5.752690410614013,
      "epoch": 2.30361030478955,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044685703026595266,
      "loss": 5.4295,
      "mean_token_accuracy": 0.17730269879102706,
      "num_tokens": 52685673.0,
      "step": 25395
    },
    {
      "entropy": 5.726068305969238,
      "epoch": 2.3040638606676342,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044683614428665314,
      "loss": 5.438,
      "mean_token_accuracy": 0.17081462889909743,
      "num_tokens": 52696696.0,
      "step": 25400
    },
    {
      "entropy": 5.662946081161499,
      "epoch": 2.3045174165457185,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044681525475373843,
      "loss": 5.4023,
      "mean_token_accuracy": 0.1720178857445717,
      "num_tokens": 52707662.0,
      "step": 25405
    },
    {
      "entropy": 5.696346282958984,
      "epoch": 2.3049709724238028,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004467943616676403,
      "loss": 5.3265,
      "mean_token_accuracy": 0.1834930568933487,
      "num_tokens": 52718438.0,
      "step": 25410
    },
    {
      "entropy": 5.676717281341553,
      "epoch": 2.3054245283018866,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004467734650287911,
      "loss": 5.3324,
      "mean_token_accuracy": 0.17604068368673326,
      "num_tokens": 52729267.0,
      "step": 25415
    },
    {
      "entropy": 5.611058855056763,
      "epoch": 2.305878084179971,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004467525648376229,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18914227783679963,
      "num_tokens": 52738754.0,
      "step": 25420
    },
    {
      "entropy": 5.66146183013916,
      "epoch": 2.306331640058055,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004467316610945679,
      "loss": 5.3443,
      "mean_token_accuracy": 0.18288354575634003,
      "num_tokens": 52749086.0,
      "step": 25425
    },
    {
      "entropy": 5.755763673782349,
      "epoch": 2.3067851959361394,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044671075380005825,
      "loss": 5.4373,
      "mean_token_accuracy": 0.1732671871781349,
      "num_tokens": 52758961.0,
      "step": 25430
    },
    {
      "entropy": 5.72589955329895,
      "epoch": 2.3072387518142237,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004466898429545266,
      "loss": 5.3613,
      "mean_token_accuracy": 0.182387176156044,
      "num_tokens": 52769447.0,
      "step": 25435
    },
    {
      "entropy": 5.7280481338500975,
      "epoch": 2.3076923076923075,
      "grad_norm": 2.296875,
      "learning_rate": 0.00044666892855840513,
      "loss": 5.4533,
      "mean_token_accuracy": 0.17027648240327836,
      "num_tokens": 52780306.0,
      "step": 25440
    },
    {
      "entropy": 5.669167566299438,
      "epoch": 2.3081458635703918,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004466480106121266,
      "loss": 5.3373,
      "mean_token_accuracy": 0.1747964769601822,
      "num_tokens": 52791263.0,
      "step": 25445
    },
    {
      "entropy": 5.70522894859314,
      "epoch": 2.308599419448476,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004466270891161234,
      "loss": 5.3551,
      "mean_token_accuracy": 0.17168041914701462,
      "num_tokens": 52802194.0,
      "step": 25450
    },
    {
      "entropy": 5.666455459594727,
      "epoch": 2.3090529753265603,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044660616407082827,
      "loss": 5.3409,
      "mean_token_accuracy": 0.18484747260808945,
      "num_tokens": 52812391.0,
      "step": 25455
    },
    {
      "entropy": 5.721006202697754,
      "epoch": 2.3095065312046446,
      "grad_norm": 2.578125,
      "learning_rate": 0.00044658523547667385,
      "loss": 5.3073,
      "mean_token_accuracy": 0.17936608344316482,
      "num_tokens": 52823540.0,
      "step": 25460
    },
    {
      "entropy": 5.617044353485108,
      "epoch": 2.309960087082729,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004465643033340931,
      "loss": 5.2508,
      "mean_token_accuracy": 0.18803484588861466,
      "num_tokens": 52833238.0,
      "step": 25465
    },
    {
      "entropy": 5.7320514678955075,
      "epoch": 2.3104136429608126,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044654336764351886,
      "loss": 5.4229,
      "mean_token_accuracy": 0.17104258090257646,
      "num_tokens": 52844510.0,
      "step": 25470
    },
    {
      "entropy": 5.745517301559448,
      "epoch": 2.310867198838897,
      "grad_norm": 2.25,
      "learning_rate": 0.00044652242840538406,
      "loss": 5.3898,
      "mean_token_accuracy": 0.18043490648269653,
      "num_tokens": 52856311.0,
      "step": 25475
    },
    {
      "entropy": 5.632608461380005,
      "epoch": 2.311320754716981,
      "grad_norm": 2.5,
      "learning_rate": 0.0004465014856201218,
      "loss": 5.2903,
      "mean_token_accuracy": 0.17971145808696748,
      "num_tokens": 52866295.0,
      "step": 25480
    },
    {
      "entropy": 5.6771790981292725,
      "epoch": 2.3117743105950654,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000446480539288165,
      "loss": 5.3591,
      "mean_token_accuracy": 0.175899700820446,
      "num_tokens": 52876818.0,
      "step": 25485
    },
    {
      "entropy": 5.657790946960449,
      "epoch": 2.3122278664731493,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004464595894099469,
      "loss": 5.2929,
      "mean_token_accuracy": 0.18111017048358918,
      "num_tokens": 52886173.0,
      "step": 25490
    },
    {
      "entropy": 5.598279762268066,
      "epoch": 2.3126814223512335,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044643863598590094,
      "loss": 5.2839,
      "mean_token_accuracy": 0.17968489676713945,
      "num_tokens": 52895875.0,
      "step": 25495
    },
    {
      "entropy": 5.699652242660522,
      "epoch": 2.313134978229318,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00044641767901646025,
      "loss": 5.4214,
      "mean_token_accuracy": 0.17255652099847793,
      "num_tokens": 52906465.0,
      "step": 25500
    },
    {
      "entropy": 5.758817481994629,
      "epoch": 2.313588534107402,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004463967185020583,
      "loss": 5.3424,
      "mean_token_accuracy": 0.18042772710323335,
      "num_tokens": 52916378.0,
      "step": 25505
    },
    {
      "entropy": 5.70608434677124,
      "epoch": 2.3140420899854863,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004463757544431285,
      "loss": 5.3951,
      "mean_token_accuracy": 0.1806505635380745,
      "num_tokens": 52926698.0,
      "step": 25510
    },
    {
      "entropy": 5.7357958316802975,
      "epoch": 2.3144956458635706,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004463547868401044,
      "loss": 5.3518,
      "mean_token_accuracy": 0.1743541181087494,
      "num_tokens": 52936469.0,
      "step": 25515
    },
    {
      "entropy": 5.655531024932861,
      "epoch": 2.3149492017416544,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044633381569341964,
      "loss": 5.366,
      "mean_token_accuracy": 0.1736176133155823,
      "num_tokens": 52947006.0,
      "step": 25520
    },
    {
      "entropy": 5.578104400634766,
      "epoch": 2.3154027576197387,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004463128410035079,
      "loss": 5.2305,
      "mean_token_accuracy": 0.18404255211353301,
      "num_tokens": 52956663.0,
      "step": 25525
    },
    {
      "entropy": 5.671026802062988,
      "epoch": 2.315856313497823,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044629186277080307,
      "loss": 5.2679,
      "mean_token_accuracy": 0.18056537657976152,
      "num_tokens": 52967383.0,
      "step": 25530
    },
    {
      "entropy": 5.612407207489014,
      "epoch": 2.316309869375907,
      "grad_norm": 7.375,
      "learning_rate": 0.00044627088099573884,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18521017134189605,
      "num_tokens": 52976567.0,
      "step": 25535
    },
    {
      "entropy": 5.639183521270752,
      "epoch": 2.3167634252539915,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044624989567874907,
      "loss": 5.351,
      "mean_token_accuracy": 0.1774131327867508,
      "num_tokens": 52986159.0,
      "step": 25540
    },
    {
      "entropy": 5.732939624786377,
      "epoch": 2.3172169811320753,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044622890682026787,
      "loss": 5.3819,
      "mean_token_accuracy": 0.17360736727714537,
      "num_tokens": 52996028.0,
      "step": 25545
    },
    {
      "entropy": 5.764906501770019,
      "epoch": 2.3176705370101596,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004462079144207292,
      "loss": 5.3829,
      "mean_token_accuracy": 0.17382852882146835,
      "num_tokens": 53006446.0,
      "step": 25550
    },
    {
      "entropy": 5.7415858745574955,
      "epoch": 2.318124092888244,
      "grad_norm": 2.515625,
      "learning_rate": 0.00044618691848056726,
      "loss": 5.3386,
      "mean_token_accuracy": 0.17464514523744584,
      "num_tokens": 53016497.0,
      "step": 25555
    },
    {
      "entropy": 5.6962419033050535,
      "epoch": 2.318577648766328,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004461659190002162,
      "loss": 5.4344,
      "mean_token_accuracy": 0.17665442377328872,
      "num_tokens": 53027811.0,
      "step": 25560
    },
    {
      "entropy": 5.678946208953858,
      "epoch": 2.3190312046444124,
      "grad_norm": 1.625,
      "learning_rate": 0.00044614491598011027,
      "loss": 5.3386,
      "mean_token_accuracy": 0.17330001741647721,
      "num_tokens": 53038206.0,
      "step": 25565
    },
    {
      "entropy": 5.719795799255371,
      "epoch": 2.319484760522496,
      "grad_norm": 2.0,
      "learning_rate": 0.0004461239094206839,
      "loss": 5.3928,
      "mean_token_accuracy": 0.18025309592485428,
      "num_tokens": 53047974.0,
      "step": 25570
    },
    {
      "entropy": 5.773519372940063,
      "epoch": 2.3199383164005805,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044610289932237154,
      "loss": 5.3977,
      "mean_token_accuracy": 0.18236434012651442,
      "num_tokens": 53057904.0,
      "step": 25575
    },
    {
      "entropy": 5.621632242202759,
      "epoch": 2.3203918722786647,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044608188568560756,
      "loss": 5.3223,
      "mean_token_accuracy": 0.1801011383533478,
      "num_tokens": 53068447.0,
      "step": 25580
    },
    {
      "entropy": 5.675791692733765,
      "epoch": 2.320845428156749,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004460608685108266,
      "loss": 5.366,
      "mean_token_accuracy": 0.17158274799585344,
      "num_tokens": 53079369.0,
      "step": 25585
    },
    {
      "entropy": 5.797316360473633,
      "epoch": 2.3212989840348333,
      "grad_norm": 2.21875,
      "learning_rate": 0.00044603984779846334,
      "loss": 5.4479,
      "mean_token_accuracy": 0.16806902289390563,
      "num_tokens": 53090794.0,
      "step": 25590
    },
    {
      "entropy": 5.662680721282959,
      "epoch": 2.321752539912917,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004460188235489524,
      "loss": 5.2503,
      "mean_token_accuracy": 0.18019117414951324,
      "num_tokens": 53100802.0,
      "step": 25595
    },
    {
      "entropy": 5.619372415542602,
      "epoch": 2.3222060957910013,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004459977957627285,
      "loss": 5.3989,
      "mean_token_accuracy": 0.17282214611768723,
      "num_tokens": 53111950.0,
      "step": 25600
    },
    {
      "entropy": 5.5743608474731445,
      "epoch": 2.3226596516690856,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00044597676444022683,
      "loss": 5.2007,
      "mean_token_accuracy": 0.18227743804454805,
      "num_tokens": 53121908.0,
      "step": 25605
    },
    {
      "entropy": 5.661539030075073,
      "epoch": 2.32311320754717,
      "grad_norm": 2.703125,
      "learning_rate": 0.00044595572958188194,
      "loss": 5.355,
      "mean_token_accuracy": 0.17405309528112411,
      "num_tokens": 53133065.0,
      "step": 25610
    },
    {
      "entropy": 5.6267719745635985,
      "epoch": 2.323566763425254,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004459346911881291,
      "loss": 5.2874,
      "mean_token_accuracy": 0.17967332005500794,
      "num_tokens": 53144135.0,
      "step": 25615
    },
    {
      "entropy": 5.711835098266602,
      "epoch": 2.3240203193033384,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004459136492594032,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18749150782823562,
      "num_tokens": 53153916.0,
      "step": 25620
    },
    {
      "entropy": 5.579347896575928,
      "epoch": 2.3244738751814222,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004458926037961394,
      "loss": 5.2404,
      "mean_token_accuracy": 0.1911630466580391,
      "num_tokens": 53163937.0,
      "step": 25625
    },
    {
      "entropy": 5.692760610580445,
      "epoch": 2.3249274310595065,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004458715547987732,
      "loss": 5.4033,
      "mean_token_accuracy": 0.17148444056510925,
      "num_tokens": 53174327.0,
      "step": 25630
    },
    {
      "entropy": 5.757743263244629,
      "epoch": 2.3253809869375908,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004458505022677396,
      "loss": 5.3539,
      "mean_token_accuracy": 0.1720525547862053,
      "num_tokens": 53185946.0,
      "step": 25635
    },
    {
      "entropy": 5.644641733169555,
      "epoch": 2.325834542815675,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004458294462034741,
      "loss": 5.3127,
      "mean_token_accuracy": 0.1801527038216591,
      "num_tokens": 53195944.0,
      "step": 25640
    },
    {
      "entropy": 5.634994697570801,
      "epoch": 2.326288098693759,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004458083866064121,
      "loss": 5.3582,
      "mean_token_accuracy": 0.17907659858465194,
      "num_tokens": 53206054.0,
      "step": 25645
    },
    {
      "entropy": 5.695556688308716,
      "epoch": 2.326741654571843,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004457873234769891,
      "loss": 5.35,
      "mean_token_accuracy": 0.17788317501544954,
      "num_tokens": 53216923.0,
      "step": 25650
    },
    {
      "entropy": 5.661757230758667,
      "epoch": 2.3271952104499274,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044576625681564086,
      "loss": 5.3454,
      "mean_token_accuracy": 0.17680461704730988,
      "num_tokens": 53226788.0,
      "step": 25655
    },
    {
      "entropy": 5.675486898422241,
      "epoch": 2.3276487663280117,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044574518662280276,
      "loss": 5.356,
      "mean_token_accuracy": 0.1776188224554062,
      "num_tokens": 53237573.0,
      "step": 25660
    },
    {
      "entropy": 5.730041742324829,
      "epoch": 2.328102322206096,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004457241128989108,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17566130310297012,
      "num_tokens": 53247724.0,
      "step": 25665
    },
    {
      "entropy": 5.684921598434448,
      "epoch": 2.32855587808418,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004457030356444006,
      "loss": 5.3324,
      "mean_token_accuracy": 0.17760881632566453,
      "num_tokens": 53258774.0,
      "step": 25670
    },
    {
      "entropy": 5.649406671524048,
      "epoch": 2.329009433962264,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004456819548597082,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1793532580137253,
      "num_tokens": 53268550.0,
      "step": 25675
    },
    {
      "entropy": 5.559152984619141,
      "epoch": 2.3294629898403483,
      "grad_norm": 1.953125,
      "learning_rate": 0.00044566087054526946,
      "loss": 5.2786,
      "mean_token_accuracy": 0.18542591631412506,
      "num_tokens": 53277975.0,
      "step": 25680
    },
    {
      "entropy": 5.7486522674560545,
      "epoch": 2.3299165457184325,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004456397827015204,
      "loss": 5.433,
      "mean_token_accuracy": 0.17159474194049834,
      "num_tokens": 53287939.0,
      "step": 25685
    },
    {
      "entropy": 5.725204706192017,
      "epoch": 2.330370101596517,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004456186913288971,
      "loss": 5.377,
      "mean_token_accuracy": 0.17900461107492446,
      "num_tokens": 53298355.0,
      "step": 25690
    },
    {
      "entropy": 5.709845733642578,
      "epoch": 2.3308236574746006,
      "grad_norm": 1.75,
      "learning_rate": 0.0004455975964278359,
      "loss": 5.4477,
      "mean_token_accuracy": 0.17348464131355285,
      "num_tokens": 53308513.0,
      "step": 25695
    },
    {
      "entropy": 5.744960165023803,
      "epoch": 2.331277213352685,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044557649799877285,
      "loss": 5.3659,
      "mean_token_accuracy": 0.177780482172966,
      "num_tokens": 53318249.0,
      "step": 25700
    },
    {
      "entropy": 5.616604375839233,
      "epoch": 2.331730769230769,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004455553960421443,
      "loss": 5.3165,
      "mean_token_accuracy": 0.1798711434006691,
      "num_tokens": 53328460.0,
      "step": 25705
    },
    {
      "entropy": 5.69071216583252,
      "epoch": 2.3321843251088534,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00044553429055838685,
      "loss": 5.3445,
      "mean_token_accuracy": 0.18611115962266922,
      "num_tokens": 53338984.0,
      "step": 25710
    },
    {
      "entropy": 5.7148384094238285,
      "epoch": 2.3326378809869377,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044551318154793676,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17601377964019777,
      "num_tokens": 53348092.0,
      "step": 25715
    },
    {
      "entropy": 5.61733078956604,
      "epoch": 2.333091436865022,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044549206901123056,
      "loss": 5.286,
      "mean_token_accuracy": 0.1823653444647789,
      "num_tokens": 53357343.0,
      "step": 25720
    },
    {
      "entropy": 5.636440324783325,
      "epoch": 2.333544992743106,
      "grad_norm": 3.234375,
      "learning_rate": 0.000445470952948705,
      "loss": 5.243,
      "mean_token_accuracy": 0.18208125680685044,
      "num_tokens": 53367343.0,
      "step": 25725
    },
    {
      "entropy": 5.764646053314209,
      "epoch": 2.33399854862119,
      "grad_norm": 2.328125,
      "learning_rate": 0.00044544983336079665,
      "loss": 5.4768,
      "mean_token_accuracy": 0.16732360422611237,
      "num_tokens": 53379093.0,
      "step": 25730
    },
    {
      "entropy": 5.696083498001099,
      "epoch": 2.3344521044992743,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004454287102479423,
      "loss": 5.3081,
      "mean_token_accuracy": 0.1795162469148636,
      "num_tokens": 53388883.0,
      "step": 25735
    },
    {
      "entropy": 5.670311355590821,
      "epoch": 2.3349056603773586,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004454075836105788,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1845689982175827,
      "num_tokens": 53399393.0,
      "step": 25740
    },
    {
      "entropy": 5.634166097640991,
      "epoch": 2.335359216255443,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044538645344914307,
      "loss": 5.2654,
      "mean_token_accuracy": 0.18457259982824326,
      "num_tokens": 53408545.0,
      "step": 25745
    },
    {
      "entropy": 5.68466010093689,
      "epoch": 2.3358127721335267,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004453653197640721,
      "loss": 5.3492,
      "mean_token_accuracy": 0.178889898955822,
      "num_tokens": 53418325.0,
      "step": 25750
    },
    {
      "entropy": 5.686120891571045,
      "epoch": 2.336266328011611,
      "grad_norm": 1.625,
      "learning_rate": 0.00044534418255580286,
      "loss": 5.3493,
      "mean_token_accuracy": 0.17127428501844405,
      "num_tokens": 53427984.0,
      "step": 25755
    },
    {
      "entropy": 5.719752168655395,
      "epoch": 2.336719883889695,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044532304182477245,
      "loss": 5.4248,
      "mean_token_accuracy": 0.16934796720743178,
      "num_tokens": 53438765.0,
      "step": 25760
    },
    {
      "entropy": 5.723668575286865,
      "epoch": 2.3371734397677795,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004453018975714182,
      "loss": 5.3442,
      "mean_token_accuracy": 0.18141748607158661,
      "num_tokens": 53448792.0,
      "step": 25765
    },
    {
      "entropy": 5.715279197692871,
      "epoch": 2.3376269956458637,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004452807497961773,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17189199030399321,
      "num_tokens": 53459334.0,
      "step": 25770
    },
    {
      "entropy": 5.689843416213989,
      "epoch": 2.3380805515239476,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044525959849948704,
      "loss": 5.4386,
      "mean_token_accuracy": 0.17335410118103028,
      "num_tokens": 53469841.0,
      "step": 25775
    },
    {
      "entropy": 5.709191751480103,
      "epoch": 2.338534107402032,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044523844368178503,
      "loss": 5.3782,
      "mean_token_accuracy": 0.1735709086060524,
      "num_tokens": 53480660.0,
      "step": 25780
    },
    {
      "entropy": 5.710656547546387,
      "epoch": 2.338987663280116,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004452172853435085,
      "loss": 5.26,
      "mean_token_accuracy": 0.18140392303466796,
      "num_tokens": 53490276.0,
      "step": 25785
    },
    {
      "entropy": 5.566821908950805,
      "epoch": 2.3394412191582004,
      "grad_norm": 1.625,
      "learning_rate": 0.0004451961234850953,
      "loss": 5.2342,
      "mean_token_accuracy": 0.19029886424541473,
      "num_tokens": 53501100.0,
      "step": 25790
    },
    {
      "entropy": 5.642734956741333,
      "epoch": 2.3398947750362846,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004451749581069827,
      "loss": 5.3022,
      "mean_token_accuracy": 0.1868653878569603,
      "num_tokens": 53511680.0,
      "step": 25795
    },
    {
      "entropy": 5.68400764465332,
      "epoch": 2.3403483309143684,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004451537892096087,
      "loss": 5.4019,
      "mean_token_accuracy": 0.1814878910779953,
      "num_tokens": 53521298.0,
      "step": 25800
    },
    {
      "entropy": 5.645365524291992,
      "epoch": 2.3408018867924527,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044513261679341094,
      "loss": 5.2435,
      "mean_token_accuracy": 0.18486382812261581,
      "num_tokens": 53530413.0,
      "step": 25805
    },
    {
      "entropy": 5.609395360946655,
      "epoch": 2.341255442670537,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004451114408588273,
      "loss": 5.3295,
      "mean_token_accuracy": 0.17874400764703752,
      "num_tokens": 53540959.0,
      "step": 25810
    },
    {
      "entropy": 5.706597900390625,
      "epoch": 2.3417089985486212,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044509026140629584,
      "loss": 5.3497,
      "mean_token_accuracy": 0.1797327071428299,
      "num_tokens": 53551265.0,
      "step": 25815
    },
    {
      "entropy": 5.733025026321411,
      "epoch": 2.3421625544267055,
      "grad_norm": 3.828125,
      "learning_rate": 0.00044506907843625427,
      "loss": 5.3798,
      "mean_token_accuracy": 0.17203713208436966,
      "num_tokens": 53562666.0,
      "step": 25820
    },
    {
      "entropy": 5.6096516132354735,
      "epoch": 2.34261611030479,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044504789194914086,
      "loss": 5.4345,
      "mean_token_accuracy": 0.17227883636951447,
      "num_tokens": 53573688.0,
      "step": 25825
    },
    {
      "entropy": 5.737540674209595,
      "epoch": 2.3430696661828736,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044502670194539374,
      "loss": 5.2937,
      "mean_token_accuracy": 0.1833721160888672,
      "num_tokens": 53584055.0,
      "step": 25830
    },
    {
      "entropy": 5.5951539993286135,
      "epoch": 2.343523222060958,
      "grad_norm": 2.921875,
      "learning_rate": 0.00044500550842545104,
      "loss": 5.2104,
      "mean_token_accuracy": 0.1886206790804863,
      "num_tokens": 53593765.0,
      "step": 25835
    },
    {
      "entropy": 5.5604041576385494,
      "epoch": 2.343976777939042,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004449843113897512,
      "loss": 5.2706,
      "mean_token_accuracy": 0.18632165789604188,
      "num_tokens": 53603480.0,
      "step": 25840
    },
    {
      "entropy": 5.663894987106323,
      "epoch": 2.3444303338171264,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004449631108387325,
      "loss": 5.3048,
      "mean_token_accuracy": 0.182084758579731,
      "num_tokens": 53613797.0,
      "step": 25845
    },
    {
      "entropy": 5.642231512069702,
      "epoch": 2.34488388969521,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004449419067728333,
      "loss": 5.2761,
      "mean_token_accuracy": 0.1816807731986046,
      "num_tokens": 53623646.0,
      "step": 25850
    },
    {
      "entropy": 5.604572296142578,
      "epoch": 2.3453374455732945,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044492069919249206,
      "loss": 5.2785,
      "mean_token_accuracy": 0.1795975685119629,
      "num_tokens": 53634118.0,
      "step": 25855
    },
    {
      "entropy": 5.622942209243774,
      "epoch": 2.3457910014513788,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004448994880981476,
      "loss": 5.3126,
      "mean_token_accuracy": 0.1765262022614479,
      "num_tokens": 53644370.0,
      "step": 25860
    },
    {
      "entropy": 5.733655881881714,
      "epoch": 2.346244557329463,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044487827349023833,
      "loss": 5.3799,
      "mean_token_accuracy": 0.1713085412979126,
      "num_tokens": 53655560.0,
      "step": 25865
    },
    {
      "entropy": 5.718778514862061,
      "epoch": 2.3466981132075473,
      "grad_norm": 2.21875,
      "learning_rate": 0.00044485705536920315,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1733272299170494,
      "num_tokens": 53666401.0,
      "step": 25870
    },
    {
      "entropy": 5.757143545150757,
      "epoch": 2.3471516690856316,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044483583373548075,
      "loss": 5.4034,
      "mean_token_accuracy": 0.18542349338531494,
      "num_tokens": 53676133.0,
      "step": 25875
    },
    {
      "entropy": 5.621232318878174,
      "epoch": 2.3476052249637154,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004448146085895101,
      "loss": 5.3245,
      "mean_token_accuracy": 0.17809472978115082,
      "num_tokens": 53687683.0,
      "step": 25880
    },
    {
      "entropy": 5.669129180908203,
      "epoch": 2.3480587808417996,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004447933799317299,
      "loss": 5.3958,
      "mean_token_accuracy": 0.17556769996881486,
      "num_tokens": 53698343.0,
      "step": 25885
    },
    {
      "entropy": 5.713808679580689,
      "epoch": 2.348512336719884,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004447721477625795,
      "loss": 5.3282,
      "mean_token_accuracy": 0.1772194802761078,
      "num_tokens": 53709014.0,
      "step": 25890
    },
    {
      "entropy": 5.612720537185669,
      "epoch": 2.348965892597968,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044475091208249774,
      "loss": 5.3421,
      "mean_token_accuracy": 0.17752332091331482,
      "num_tokens": 53718838.0,
      "step": 25895
    },
    {
      "entropy": 5.60744080543518,
      "epoch": 2.3494194484760524,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004447296728919239,
      "loss": 5.2729,
      "mean_token_accuracy": 0.18087218403816224,
      "num_tokens": 53728526.0,
      "step": 25900
    },
    {
      "entropy": 5.701401662826538,
      "epoch": 2.3498730043541363,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004447084301912972,
      "loss": 5.4109,
      "mean_token_accuracy": 0.17571754306554793,
      "num_tokens": 53739261.0,
      "step": 25905
    },
    {
      "entropy": 5.706050682067871,
      "epoch": 2.3503265602322205,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004446871839810568,
      "loss": 5.382,
      "mean_token_accuracy": 0.1787704199552536,
      "num_tokens": 53749599.0,
      "step": 25910
    },
    {
      "entropy": 5.662590169906617,
      "epoch": 2.350780116110305,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00044466593426164224,
      "loss": 5.3595,
      "mean_token_accuracy": 0.17721371948719025,
      "num_tokens": 53759898.0,
      "step": 25915
    },
    {
      "entropy": 5.68318657875061,
      "epoch": 2.351233671988389,
      "grad_norm": 2.078125,
      "learning_rate": 0.000444644681033493,
      "loss": 5.3092,
      "mean_token_accuracy": 0.1742246314883232,
      "num_tokens": 53770561.0,
      "step": 25920
    },
    {
      "entropy": 5.657411193847656,
      "epoch": 2.3516872278664733,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00044462342429704845,
      "loss": 5.3035,
      "mean_token_accuracy": 0.17948016226291658,
      "num_tokens": 53780408.0,
      "step": 25925
    },
    {
      "entropy": 5.555493116378784,
      "epoch": 2.352140783744557,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044460216405274827,
      "loss": 5.2527,
      "mean_token_accuracy": 0.19087414145469667,
      "num_tokens": 53789125.0,
      "step": 25930
    },
    {
      "entropy": 5.666281032562256,
      "epoch": 2.3525943396226414,
      "grad_norm": 2.3125,
      "learning_rate": 0.00044458090030103216,
      "loss": 5.2727,
      "mean_token_accuracy": 0.18575793653726577,
      "num_tokens": 53799164.0,
      "step": 25935
    },
    {
      "entropy": 5.683192014694214,
      "epoch": 2.3530478955007257,
      "grad_norm": 2.3125,
      "learning_rate": 0.00044455963304233976,
      "loss": 5.2891,
      "mean_token_accuracy": 0.18120076060295104,
      "num_tokens": 53810710.0,
      "step": 25940
    },
    {
      "entropy": 5.592425203323364,
      "epoch": 2.35350145137881,
      "grad_norm": 2.453125,
      "learning_rate": 0.000444538362277111,
      "loss": 5.301,
      "mean_token_accuracy": 0.17587408274412156,
      "num_tokens": 53820692.0,
      "step": 25945
    },
    {
      "entropy": 5.7308629035949705,
      "epoch": 2.353955007256894,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004445170880057857,
      "loss": 5.3701,
      "mean_token_accuracy": 0.1761684775352478,
      "num_tokens": 53831082.0,
      "step": 25950
    },
    {
      "entropy": 5.7388263702392575,
      "epoch": 2.354408563134978,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00044449581022880375,
      "loss": 5.4061,
      "mean_token_accuracy": 0.16705995351076125,
      "num_tokens": 53842602.0,
      "step": 25955
    },
    {
      "entropy": 5.675936079025268,
      "epoch": 2.3548621190130623,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004444745289466053,
      "loss": 5.3399,
      "mean_token_accuracy": 0.17842220067977904,
      "num_tokens": 53853784.0,
      "step": 25960
    },
    {
      "entropy": 5.65016016960144,
      "epoch": 2.3553156748911466,
      "grad_norm": 2.390625,
      "learning_rate": 0.00044445324415963043,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18788180947303773,
      "num_tokens": 53863526.0,
      "step": 25965
    },
    {
      "entropy": 5.580548477172852,
      "epoch": 2.355769230769231,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004444319558683192,
      "loss": 5.3311,
      "mean_token_accuracy": 0.18160877227783204,
      "num_tokens": 53874594.0,
      "step": 25970
    },
    {
      "entropy": 5.618749332427979,
      "epoch": 2.356222786647315,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044441066407311205,
      "loss": 5.3494,
      "mean_token_accuracy": 0.17417338192462922,
      "num_tokens": 53884771.0,
      "step": 25975
    },
    {
      "entropy": 5.719550085067749,
      "epoch": 2.3566763425253994,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004443893687744492,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17956369817256929,
      "num_tokens": 53895181.0,
      "step": 25980
    },
    {
      "entropy": 5.594526529312134,
      "epoch": 2.357129898403483,
      "grad_norm": 2.84375,
      "learning_rate": 0.00044436806997277096,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18869947344064714,
      "num_tokens": 53904153.0,
      "step": 25985
    },
    {
      "entropy": 5.654611682891845,
      "epoch": 2.3575834542815675,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044434676766851787,
      "loss": 5.3105,
      "mean_token_accuracy": 0.17836464643478395,
      "num_tokens": 53914064.0,
      "step": 25990
    },
    {
      "entropy": 5.716101741790771,
      "epoch": 2.3580370101596517,
      "grad_norm": 3.75,
      "learning_rate": 0.00044432546186213054,
      "loss": 5.3359,
      "mean_token_accuracy": 0.1759283050894737,
      "num_tokens": 53924240.0,
      "step": 25995
    },
    {
      "entropy": 5.735394525527954,
      "epoch": 2.358490566037736,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044430415255404944,
      "loss": 5.4054,
      "mean_token_accuracy": 0.18011405169963837,
      "num_tokens": 53935395.0,
      "step": 26000
    },
    {
      "entropy": 5.635989141464234,
      "epoch": 2.35894412191582,
      "grad_norm": 3.046875,
      "learning_rate": 0.00044428283974471533,
      "loss": 5.315,
      "mean_token_accuracy": 0.17865053117275237,
      "num_tokens": 53945768.0,
      "step": 26005
    },
    {
      "entropy": 5.584124565124512,
      "epoch": 2.359397677793904,
      "grad_norm": 2.6875,
      "learning_rate": 0.000444261523434569,
      "loss": 5.2946,
      "mean_token_accuracy": 0.18513273447752,
      "num_tokens": 53956467.0,
      "step": 26010
    },
    {
      "entropy": 5.61675534248352,
      "epoch": 2.3598512336719883,
      "grad_norm": 3.46875,
      "learning_rate": 0.00044424020362405126,
      "loss": 5.2403,
      "mean_token_accuracy": 0.18453543931245803,
      "num_tokens": 53967587.0,
      "step": 26015
    },
    {
      "entropy": 5.609177732467652,
      "epoch": 2.3603047895500726,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004442188803136029,
      "loss": 5.2743,
      "mean_token_accuracy": 0.1832869380712509,
      "num_tokens": 53977691.0,
      "step": 26020
    },
    {
      "entropy": 5.590100908279419,
      "epoch": 2.360758345428157,
      "grad_norm": 2.09375,
      "learning_rate": 0.000444197553503665,
      "loss": 5.3821,
      "mean_token_accuracy": 0.17125103026628494,
      "num_tokens": 53987608.0,
      "step": 26025
    },
    {
      "entropy": 5.74505205154419,
      "epoch": 2.361211901306241,
      "grad_norm": 3.53125,
      "learning_rate": 0.00044417622319467857,
      "loss": 5.4392,
      "mean_token_accuracy": 0.17236627340316774,
      "num_tokens": 53998010.0,
      "step": 26030
    },
    {
      "entropy": 5.690523958206176,
      "epoch": 2.361665457184325,
      "grad_norm": 1.96875,
      "learning_rate": 0.00044415488938708475,
      "loss": 5.3319,
      "mean_token_accuracy": 0.17985129058361055,
      "num_tokens": 54009852.0,
      "step": 26035
    },
    {
      "entropy": 5.687312746047974,
      "epoch": 2.3621190130624092,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004441335520813247,
      "loss": 5.2878,
      "mean_token_accuracy": 0.177458293735981,
      "num_tokens": 54019635.0,
      "step": 26040
    },
    {
      "entropy": 5.674623727798462,
      "epoch": 2.3625725689404935,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044411221127783966,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17884511947631837,
      "num_tokens": 54030046.0,
      "step": 26045
    },
    {
      "entropy": 5.674948501586914,
      "epoch": 2.3630261248185778,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004440908669770711,
      "loss": 5.3061,
      "mean_token_accuracy": 0.17782025933265685,
      "num_tokens": 54041197.0,
      "step": 26050
    },
    {
      "entropy": 5.704526805877686,
      "epoch": 2.3634796806966616,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004440695191794602,
      "loss": 5.2896,
      "mean_token_accuracy": 0.18040791898965836,
      "num_tokens": 54051592.0,
      "step": 26055
    },
    {
      "entropy": 5.666587400436401,
      "epoch": 2.363933236574746,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004440481678854487,
      "loss": 5.3741,
      "mean_token_accuracy": 0.1732002690434456,
      "num_tokens": 54062893.0,
      "step": 26060
    },
    {
      "entropy": 5.691020774841308,
      "epoch": 2.36438679245283,
      "grad_norm": 2.140625,
      "learning_rate": 0.00044402681309547786,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17563274800777434,
      "num_tokens": 54073651.0,
      "step": 26065
    },
    {
      "entropy": 5.669061517715454,
      "epoch": 2.3648403483309144,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044400545480998956,
      "loss": 5.3099,
      "mean_token_accuracy": 0.18526316434144974,
      "num_tokens": 54084889.0,
      "step": 26070
    },
    {
      "entropy": 5.605622148513794,
      "epoch": 2.3652939042089987,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044398409302942533,
      "loss": 5.3377,
      "mean_token_accuracy": 0.17980865091085435,
      "num_tokens": 54095724.0,
      "step": 26075
    },
    {
      "entropy": 5.612457704544068,
      "epoch": 2.365747460087083,
      "grad_norm": 1.46875,
      "learning_rate": 0.000443962727754227,
      "loss": 5.2905,
      "mean_token_accuracy": 0.1772426962852478,
      "num_tokens": 54106581.0,
      "step": 26080
    },
    {
      "entropy": 5.687488698959351,
      "epoch": 2.3662010159651667,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004439413589848364,
      "loss": 5.3268,
      "mean_token_accuracy": 0.1800633504986763,
      "num_tokens": 54116074.0,
      "step": 26085
    },
    {
      "entropy": 5.696656131744385,
      "epoch": 2.366654571843251,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004439199867216955,
      "loss": 5.3632,
      "mean_token_accuracy": 0.17446572333574295,
      "num_tokens": 54126902.0,
      "step": 26090
    },
    {
      "entropy": 5.6202311515808105,
      "epoch": 2.3671081277213353,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004438986109652462,
      "loss": 5.2928,
      "mean_token_accuracy": 0.18142935633659363,
      "num_tokens": 54137554.0,
      "step": 26095
    },
    {
      "entropy": 5.632568359375,
      "epoch": 2.3675616835994195,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004438772317159306,
      "loss": 5.2838,
      "mean_token_accuracy": 0.17879415899515153,
      "num_tokens": 54148035.0,
      "step": 26100
    },
    {
      "entropy": 5.674480962753296,
      "epoch": 2.368015239477504,
      "grad_norm": 2.265625,
      "learning_rate": 0.00044385584897419075,
      "loss": 5.289,
      "mean_token_accuracy": 0.18750009536743165,
      "num_tokens": 54157417.0,
      "step": 26105
    },
    {
      "entropy": 5.709169149398804,
      "epoch": 2.3684687953555876,
      "grad_norm": 1.734375,
      "learning_rate": 0.000443834462740469,
      "loss": 5.3414,
      "mean_token_accuracy": 0.1801222860813141,
      "num_tokens": 54168353.0,
      "step": 26110
    },
    {
      "entropy": 5.638660573959351,
      "epoch": 2.368922351233672,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004438130730152074,
      "loss": 5.3889,
      "mean_token_accuracy": 0.17235763669013976,
      "num_tokens": 54180182.0,
      "step": 26115
    },
    {
      "entropy": 5.67041015625,
      "epoch": 2.369375907111756,
      "grad_norm": 2.984375,
      "learning_rate": 0.00044379167979884847,
      "loss": 5.3357,
      "mean_token_accuracy": 0.17984447181224822,
      "num_tokens": 54190973.0,
      "step": 26120
    },
    {
      "entropy": 5.686223030090332,
      "epoch": 2.3698294629898404,
      "grad_norm": 2.125,
      "learning_rate": 0.0004437702830918347,
      "loss": 5.4025,
      "mean_token_accuracy": 0.17504687160253524,
      "num_tokens": 54202077.0,
      "step": 26125
    },
    {
      "entropy": 5.652991151809692,
      "epoch": 2.3702830188679247,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004437488828946084,
      "loss": 5.3142,
      "mean_token_accuracy": 0.1767681747674942,
      "num_tokens": 54211778.0,
      "step": 26130
    },
    {
      "entropy": 5.632948160171509,
      "epoch": 2.3707365747460085,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004437274792076122,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18876284956932068,
      "num_tokens": 54220941.0,
      "step": 26135
    },
    {
      "entropy": 5.681721830368042,
      "epoch": 2.371190130624093,
      "grad_norm": 3.78125,
      "learning_rate": 0.00044370607203128874,
      "loss": 5.407,
      "mean_token_accuracy": 0.17519531697034835,
      "num_tokens": 54231602.0,
      "step": 26140
    },
    {
      "entropy": 5.667259311676025,
      "epoch": 2.371643686502177,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004436846613660807,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17905765175819396,
      "num_tokens": 54241797.0,
      "step": 26145
    },
    {
      "entropy": 5.70704836845398,
      "epoch": 2.3720972423802613,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004436632472124308,
      "loss": 5.3969,
      "mean_token_accuracy": 0.17606339752674102,
      "num_tokens": 54253085.0,
      "step": 26150
    },
    {
      "entropy": 5.803530216217041,
      "epoch": 2.3725507982583456,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004436418295707821,
      "loss": 5.3453,
      "mean_token_accuracy": 0.18087163865566253,
      "num_tokens": 54262164.0,
      "step": 26155
    },
    {
      "entropy": 5.649451971054077,
      "epoch": 2.3730043541364294,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004436204084415774,
      "loss": 5.3439,
      "mean_token_accuracy": 0.18062433302402497,
      "num_tokens": 54271922.0,
      "step": 26160
    },
    {
      "entropy": 5.673418378829956,
      "epoch": 2.3734579100145137,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004435989838252596,
      "loss": 5.2993,
      "mean_token_accuracy": 0.18081478625535966,
      "num_tokens": 54281738.0,
      "step": 26165
    },
    {
      "entropy": 5.668488931655884,
      "epoch": 2.373911465892598,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004435775557222719,
      "loss": 5.3372,
      "mean_token_accuracy": 0.1787610322237015,
      "num_tokens": 54292252.0,
      "step": 26170
    },
    {
      "entropy": 5.6756861209869385,
      "epoch": 2.374365021770682,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044355612413305736,
      "loss": 5.3688,
      "mean_token_accuracy": 0.1776259645819664,
      "num_tokens": 54302518.0,
      "step": 26175
    },
    {
      "entropy": 5.675545263290405,
      "epoch": 2.3748185776487665,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004435346890580593,
      "loss": 5.3306,
      "mean_token_accuracy": 0.18143008202314376,
      "num_tokens": 54312917.0,
      "step": 26180
    },
    {
      "entropy": 5.674512100219727,
      "epoch": 2.3752721335268507,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044351325049772085,
      "loss": 5.2965,
      "mean_token_accuracy": 0.1837981313467026,
      "num_tokens": 54322552.0,
      "step": 26185
    },
    {
      "entropy": 5.631934452056885,
      "epoch": 2.3757256894049346,
      "grad_norm": 1.875,
      "learning_rate": 0.00044349180845248536,
      "loss": 5.331,
      "mean_token_accuracy": 0.17998016029596328,
      "num_tokens": 54332917.0,
      "step": 26190
    },
    {
      "entropy": 5.580878591537475,
      "epoch": 2.376179245283019,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004434703629227964,
      "loss": 5.3024,
      "mean_token_accuracy": 0.18243526071310043,
      "num_tokens": 54342791.0,
      "step": 26195
    },
    {
      "entropy": 5.614203405380249,
      "epoch": 2.376632801161103,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004434489139090975,
      "loss": 5.2012,
      "mean_token_accuracy": 0.18913818299770355,
      "num_tokens": 54352821.0,
      "step": 26200
    },
    {
      "entropy": 5.7759099960327145,
      "epoch": 2.3770863570391874,
      "grad_norm": 2.265625,
      "learning_rate": 0.00044342746141183206,
      "loss": 5.4041,
      "mean_token_accuracy": 0.17843062728643416,
      "num_tokens": 54363095.0,
      "step": 26205
    },
    {
      "entropy": 5.65583119392395,
      "epoch": 2.377539912917271,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004434060054314438,
      "loss": 5.2966,
      "mean_token_accuracy": 0.1792072683572769,
      "num_tokens": 54373841.0,
      "step": 26210
    },
    {
      "entropy": 5.573833084106445,
      "epoch": 2.3779934687953554,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044338454596837637,
      "loss": 5.2625,
      "mean_token_accuracy": 0.18316925764083863,
      "num_tokens": 54385012.0,
      "step": 26215
    },
    {
      "entropy": 5.610978889465332,
      "epoch": 2.3784470246734397,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004433630830230737,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1795179069042206,
      "num_tokens": 54395298.0,
      "step": 26220
    },
    {
      "entropy": 5.624135732650757,
      "epoch": 2.378900580551524,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004433416165959795,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1767999440431595,
      "num_tokens": 54405752.0,
      "step": 26225
    },
    {
      "entropy": 5.698334121704102,
      "epoch": 2.3793541364296082,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004433201466875378,
      "loss": 5.3318,
      "mean_token_accuracy": 0.1739984408020973,
      "num_tokens": 54416607.0,
      "step": 26230
    },
    {
      "entropy": 5.704953670501709,
      "epoch": 2.3798076923076925,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004432986732981926,
      "loss": 5.3046,
      "mean_token_accuracy": 0.1883317619562149,
      "num_tokens": 54426873.0,
      "step": 26235
    },
    {
      "entropy": 5.61703314781189,
      "epoch": 2.3802612481857763,
      "grad_norm": 1.609375,
      "learning_rate": 0.000443277196428388,
      "loss": 5.237,
      "mean_token_accuracy": 0.18924930542707444,
      "num_tokens": 54437611.0,
      "step": 26240
    },
    {
      "entropy": 5.701129913330078,
      "epoch": 2.3807148040638606,
      "grad_norm": 2.1875,
      "learning_rate": 0.000443255716078568,
      "loss": 5.3517,
      "mean_token_accuracy": 0.17862485945224763,
      "num_tokens": 54448992.0,
      "step": 26245
    },
    {
      "entropy": 5.662423229217529,
      "epoch": 2.381168359941945,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004432342322491769,
      "loss": 5.293,
      "mean_token_accuracy": 0.18815709799528121,
      "num_tokens": 54459172.0,
      "step": 26250
    },
    {
      "entropy": 5.577565240859985,
      "epoch": 2.381621915820029,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044321274494065903,
      "loss": 5.2756,
      "mean_token_accuracy": 0.18578815460205078,
      "num_tokens": 54469660.0,
      "step": 26255
    },
    {
      "entropy": 5.68136830329895,
      "epoch": 2.3820754716981134,
      "grad_norm": 2.0625,
      "learning_rate": 0.00044319125415345886,
      "loss": 5.381,
      "mean_token_accuracy": 0.17880837619304657,
      "num_tokens": 54479567.0,
      "step": 26260
    },
    {
      "entropy": 5.643814039230347,
      "epoch": 2.3825290275761972,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004431697598880205,
      "loss": 5.3617,
      "mean_token_accuracy": 0.17707927227020265,
      "num_tokens": 54489690.0,
      "step": 26265
    },
    {
      "entropy": 5.657520532608032,
      "epoch": 2.3829825834542815,
      "grad_norm": 2.765625,
      "learning_rate": 0.00044314826214478877,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17708388864994049,
      "num_tokens": 54500876.0,
      "step": 26270
    },
    {
      "entropy": 5.6623705387115475,
      "epoch": 2.3834361393323658,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044312676092420804,
      "loss": 5.3519,
      "mean_token_accuracy": 0.17196874767541886,
      "num_tokens": 54510988.0,
      "step": 26275
    },
    {
      "entropy": 5.675397539138794,
      "epoch": 2.38388969521045,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004431052562267231,
      "loss": 5.3057,
      "mean_token_accuracy": 0.1814639911055565,
      "num_tokens": 54522149.0,
      "step": 26280
    },
    {
      "entropy": 5.748640441894532,
      "epoch": 2.3843432510885343,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044308374805277867,
      "loss": 5.4434,
      "mean_token_accuracy": 0.1721394792199135,
      "num_tokens": 54533482.0,
      "step": 26285
    },
    {
      "entropy": 5.680763578414917,
      "epoch": 2.384796806966618,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004430622364028194,
      "loss": 5.3605,
      "mean_token_accuracy": 0.1738498717546463,
      "num_tokens": 54543058.0,
      "step": 26290
    },
    {
      "entropy": 5.6691906452178955,
      "epoch": 2.3852503628447024,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044304072127729035,
      "loss": 5.3154,
      "mean_token_accuracy": 0.18009862750768663,
      "num_tokens": 54552655.0,
      "step": 26295
    },
    {
      "entropy": 5.638441658020019,
      "epoch": 2.3857039187227866,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044301920267663633,
      "loss": 5.2575,
      "mean_token_accuracy": 0.1773529678583145,
      "num_tokens": 54562696.0,
      "step": 26300
    },
    {
      "entropy": 5.685207414627075,
      "epoch": 2.386157474600871,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004429976806013024,
      "loss": 5.2982,
      "mean_token_accuracy": 0.18799882233142853,
      "num_tokens": 54572803.0,
      "step": 26305
    },
    {
      "entropy": 5.577423477172852,
      "epoch": 2.386611030478955,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004429761550517336,
      "loss": 5.3562,
      "mean_token_accuracy": 0.178684563934803,
      "num_tokens": 54582450.0,
      "step": 26310
    },
    {
      "entropy": 5.533674049377441,
      "epoch": 2.387064586357039,
      "grad_norm": 2.21875,
      "learning_rate": 0.00044295462602837504,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1814552515745163,
      "num_tokens": 54591868.0,
      "step": 26315
    },
    {
      "entropy": 5.7030189037323,
      "epoch": 2.3875181422351233,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004429330935316721,
      "loss": 5.3088,
      "mean_token_accuracy": 0.17592508345842361,
      "num_tokens": 54601858.0,
      "step": 26320
    },
    {
      "entropy": 5.686510276794434,
      "epoch": 2.3879716981132075,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044291155756206996,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1774248704314232,
      "num_tokens": 54612243.0,
      "step": 26325
    },
    {
      "entropy": 5.635846090316773,
      "epoch": 2.388425253991292,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044289001812001404,
      "loss": 5.2509,
      "mean_token_accuracy": 0.1823895812034607,
      "num_tokens": 54622365.0,
      "step": 26330
    },
    {
      "entropy": 5.647223854064942,
      "epoch": 2.388878809869376,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004428684752059497,
      "loss": 5.2416,
      "mean_token_accuracy": 0.18745090514421464,
      "num_tokens": 54632356.0,
      "step": 26335
    },
    {
      "entropy": 5.701996326446533,
      "epoch": 2.3893323657474603,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004428469288203226,
      "loss": 5.3415,
      "mean_token_accuracy": 0.17486019879579545,
      "num_tokens": 54642976.0,
      "step": 26340
    },
    {
      "entropy": 5.760047769546508,
      "epoch": 2.389785921625544,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004428253789635783,
      "loss": 5.366,
      "mean_token_accuracy": 0.18114445358514786,
      "num_tokens": 54652597.0,
      "step": 26345
    },
    {
      "entropy": 5.655825090408325,
      "epoch": 2.3902394775036284,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004428038256361622,
      "loss": 5.4621,
      "mean_token_accuracy": 0.17219481766223907,
      "num_tokens": 54662930.0,
      "step": 26350
    },
    {
      "entropy": 5.829427719116211,
      "epoch": 2.3906930333817127,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004427822688385203,
      "loss": 5.557,
      "mean_token_accuracy": 0.17601973861455916,
      "num_tokens": 54674562.0,
      "step": 26355
    },
    {
      "entropy": 5.722427654266357,
      "epoch": 2.391146589259797,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004427607085710984,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17841463834047316,
      "num_tokens": 54684432.0,
      "step": 26360
    },
    {
      "entropy": 5.705434703826905,
      "epoch": 2.3916001451378808,
      "grad_norm": 1.375,
      "learning_rate": 0.00044273914483434226,
      "loss": 5.3367,
      "mean_token_accuracy": 0.17885835766792296,
      "num_tokens": 54694060.0,
      "step": 26365
    },
    {
      "entropy": 5.614196681976319,
      "epoch": 2.392053701015965,
      "grad_norm": 2.375,
      "learning_rate": 0.00044271757762869777,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1740858793258667,
      "num_tokens": 54704411.0,
      "step": 26370
    },
    {
      "entropy": 5.6038477420806885,
      "epoch": 2.3925072568940493,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004426960069546112,
      "loss": 5.3406,
      "mean_token_accuracy": 0.18148724883794784,
      "num_tokens": 54715185.0,
      "step": 26375
    },
    {
      "entropy": 5.653721714019776,
      "epoch": 2.3929608127721336,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004426744328125284,
      "loss": 5.2321,
      "mean_token_accuracy": 0.18761623054742813,
      "num_tokens": 54725529.0,
      "step": 26380
    },
    {
      "entropy": 5.743239068984986,
      "epoch": 2.393414368650218,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004426528552028955,
      "loss": 5.4158,
      "mean_token_accuracy": 0.1771687611937523,
      "num_tokens": 54736992.0,
      "step": 26385
    },
    {
      "entropy": 5.672323751449585,
      "epoch": 2.393867924528302,
      "grad_norm": 2.46875,
      "learning_rate": 0.00044263127412615897,
      "loss": 5.2894,
      "mean_token_accuracy": 0.18095309287309647,
      "num_tokens": 54746951.0,
      "step": 26390
    },
    {
      "entropy": 5.611057996749878,
      "epoch": 2.394321480406386,
      "grad_norm": 1.875,
      "learning_rate": 0.0004426096895827649,
      "loss": 5.2707,
      "mean_token_accuracy": 0.18685234636068343,
      "num_tokens": 54757272.0,
      "step": 26395
    },
    {
      "entropy": 5.656089496612549,
      "epoch": 2.39477503628447,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044258810157315977,
      "loss": 5.3872,
      "mean_token_accuracy": 0.17472484856843948,
      "num_tokens": 54767835.0,
      "step": 26400
    },
    {
      "entropy": 5.681936120986938,
      "epoch": 2.3952285921625545,
      "grad_norm": 1.625,
      "learning_rate": 0.00044256651009779,
      "loss": 5.3291,
      "mean_token_accuracy": 0.18453604578971863,
      "num_tokens": 54778432.0,
      "step": 26405
    },
    {
      "entropy": 5.704746437072754,
      "epoch": 2.3956821480406387,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044254491515710213,
      "loss": 5.3919,
      "mean_token_accuracy": 0.17403220385313034,
      "num_tokens": 54788582.0,
      "step": 26410
    },
    {
      "entropy": 5.708057069778443,
      "epoch": 2.3961357039187225,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004425233167515426,
      "loss": 5.3739,
      "mean_token_accuracy": 0.1805973157286644,
      "num_tokens": 54798848.0,
      "step": 26415
    },
    {
      "entropy": 5.653639888763427,
      "epoch": 2.396589259796807,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004425017148815583,
      "loss": 5.2838,
      "mean_token_accuracy": 0.1765883281826973,
      "num_tokens": 54809867.0,
      "step": 26420
    },
    {
      "entropy": 5.663675689697266,
      "epoch": 2.397042815674891,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044248010954759575,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17721291184425353,
      "num_tokens": 54819905.0,
      "step": 26425
    },
    {
      "entropy": 5.649288845062256,
      "epoch": 2.3974963715529753,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044245850075010193,
      "loss": 5.3256,
      "mean_token_accuracy": 0.18091987818479538,
      "num_tokens": 54829723.0,
      "step": 26430
    },
    {
      "entropy": 5.588828659057617,
      "epoch": 2.3979499274310596,
      "grad_norm": 2.421875,
      "learning_rate": 0.00044243688848952364,
      "loss": 5.2839,
      "mean_token_accuracy": 0.1787870094180107,
      "num_tokens": 54840008.0,
      "step": 26435
    },
    {
      "entropy": 5.665616989135742,
      "epoch": 2.398403483309144,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004424152727663078,
      "loss": 5.3282,
      "mean_token_accuracy": 0.1772574618458748,
      "num_tokens": 54851693.0,
      "step": 26440
    },
    {
      "entropy": 5.7238935947418215,
      "epoch": 2.3988570391872277,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004423936535809015,
      "loss": 5.302,
      "mean_token_accuracy": 0.18072343170642852,
      "num_tokens": 54862086.0,
      "step": 26445
    },
    {
      "entropy": 5.777760171890259,
      "epoch": 2.399310595065312,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004423720309337517,
      "loss": 5.4344,
      "mean_token_accuracy": 0.17741697728633882,
      "num_tokens": 54872883.0,
      "step": 26450
    },
    {
      "entropy": 5.676302480697632,
      "epoch": 2.3997641509433962,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044235040482530576,
      "loss": 5.3338,
      "mean_token_accuracy": 0.17483328878879548,
      "num_tokens": 54882815.0,
      "step": 26455
    },
    {
      "entropy": 5.4524318218231205,
      "epoch": 2.4002177068214805,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004423287752560108,
      "loss": 5.1434,
      "mean_token_accuracy": 0.19199773967266082,
      "num_tokens": 54892480.0,
      "step": 26460
    },
    {
      "entropy": 5.5449220657348635,
      "epoch": 2.4006712626995648,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044230714222631397,
      "loss": 5.1999,
      "mean_token_accuracy": 0.19079766273498536,
      "num_tokens": 54901607.0,
      "step": 26465
    },
    {
      "entropy": 5.803347396850586,
      "epoch": 2.4011248185776486,
      "grad_norm": 1.9375,
      "learning_rate": 0.00044228550573666294,
      "loss": 5.4849,
      "mean_token_accuracy": 0.17757042050361632,
      "num_tokens": 54913545.0,
      "step": 26470
    },
    {
      "entropy": 5.7763786792755125,
      "epoch": 2.401578374455733,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000442263865787505,
      "loss": 5.3583,
      "mean_token_accuracy": 0.17665546834468843,
      "num_tokens": 54924392.0,
      "step": 26475
    },
    {
      "entropy": 5.684337902069092,
      "epoch": 2.402031930333817,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004422422223792877,
      "loss": 5.3213,
      "mean_token_accuracy": 0.18278272747993468,
      "num_tokens": 54934114.0,
      "step": 26480
    },
    {
      "entropy": 5.713677644729614,
      "epoch": 2.4024854862119014,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004422205755124586,
      "loss": 5.3453,
      "mean_token_accuracy": 0.17602403461933136,
      "num_tokens": 54944268.0,
      "step": 26485
    },
    {
      "entropy": 5.672900581359864,
      "epoch": 2.4029390420899857,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044219892518746536,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17398566901683807,
      "num_tokens": 54954877.0,
      "step": 26490
    },
    {
      "entropy": 5.625238561630249,
      "epoch": 2.4033925979680695,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044217727140475585,
      "loss": 5.2864,
      "mean_token_accuracy": 0.18633334040641786,
      "num_tokens": 54964745.0,
      "step": 26495
    },
    {
      "entropy": 5.633251190185547,
      "epoch": 2.4038461538461537,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044215561416477764,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18022608011960983,
      "num_tokens": 54975256.0,
      "step": 26500
    },
    {
      "entropy": 5.740612411499024,
      "epoch": 2.404299709724238,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004421339534679788,
      "loss": 5.3671,
      "mean_token_accuracy": 0.17540551722049713,
      "num_tokens": 54986226.0,
      "step": 26505
    },
    {
      "entropy": 5.737144947052002,
      "epoch": 2.4047532656023223,
      "grad_norm": 2.25,
      "learning_rate": 0.0004421122893148072,
      "loss": 5.3691,
      "mean_token_accuracy": 0.16948507726192474,
      "num_tokens": 54997350.0,
      "step": 26510
    },
    {
      "entropy": 5.684903478622436,
      "epoch": 2.4052068214804065,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044209062170571086,
      "loss": 5.3254,
      "mean_token_accuracy": 0.17980141937732697,
      "num_tokens": 55007947.0,
      "step": 26515
    },
    {
      "entropy": 5.662339878082276,
      "epoch": 2.4056603773584904,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004420689506411379,
      "loss": 5.3461,
      "mean_token_accuracy": 0.18065505027770995,
      "num_tokens": 55018789.0,
      "step": 26520
    },
    {
      "entropy": 5.625288105010986,
      "epoch": 2.4061139332365746,
      "grad_norm": 1.953125,
      "learning_rate": 0.00044204727612153643,
      "loss": 5.2776,
      "mean_token_accuracy": 0.1770249292254448,
      "num_tokens": 55028934.0,
      "step": 26525
    },
    {
      "entropy": 5.658769512176514,
      "epoch": 2.406567489114659,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044202559814735475,
      "loss": 5.3004,
      "mean_token_accuracy": 0.1839599072933197,
      "num_tokens": 55039395.0,
      "step": 26530
    },
    {
      "entropy": 5.653385162353516,
      "epoch": 2.407021044992743,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004420039167190411,
      "loss": 5.2419,
      "mean_token_accuracy": 0.18224923610687255,
      "num_tokens": 55049932.0,
      "step": 26535
    },
    {
      "entropy": 5.687928199768066,
      "epoch": 2.4074746008708274,
      "grad_norm": 1.515625,
      "learning_rate": 0.000441982231837044,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17205966562032698,
      "num_tokens": 55061532.0,
      "step": 26540
    },
    {
      "entropy": 5.735758590698242,
      "epoch": 2.4079281567489117,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004419605435018117,
      "loss": 5.3815,
      "mean_token_accuracy": 0.17631963789463043,
      "num_tokens": 55072368.0,
      "step": 26545
    },
    {
      "entropy": 5.706729316711426,
      "epoch": 2.4083817126269955,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004419388517137928,
      "loss": 5.3332,
      "mean_token_accuracy": 0.17644289135932922,
      "num_tokens": 55083128.0,
      "step": 26550
    },
    {
      "entropy": 5.683570861816406,
      "epoch": 2.40883526850508,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044191715647343594,
      "loss": 5.3292,
      "mean_token_accuracy": 0.1804862439632416,
      "num_tokens": 55094071.0,
      "step": 26555
    },
    {
      "entropy": 5.6594947338104244,
      "epoch": 2.409288824383164,
      "grad_norm": 2.5,
      "learning_rate": 0.00044189545778118973,
      "loss": 5.2882,
      "mean_token_accuracy": 0.1834310546517372,
      "num_tokens": 55104504.0,
      "step": 26560
    },
    {
      "entropy": 5.6518247604370115,
      "epoch": 2.4097423802612483,
      "grad_norm": 2.234375,
      "learning_rate": 0.000441873755637503,
      "loss": 5.3257,
      "mean_token_accuracy": 0.17920529395341872,
      "num_tokens": 55115722.0,
      "step": 26565
    },
    {
      "entropy": 5.605977201461792,
      "epoch": 2.410195936139332,
      "grad_norm": 2.375,
      "learning_rate": 0.0004418520500428244,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1804247245192528,
      "num_tokens": 55126628.0,
      "step": 26570
    },
    {
      "entropy": 5.63662109375,
      "epoch": 2.4106494920174164,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044183034099760287,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18289166390895845,
      "num_tokens": 55137352.0,
      "step": 26575
    },
    {
      "entropy": 5.705312347412109,
      "epoch": 2.4111030478955007,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004418086285022874,
      "loss": 5.3148,
      "mean_token_accuracy": 0.1850166365504265,
      "num_tokens": 55147145.0,
      "step": 26580
    },
    {
      "entropy": 5.677221870422363,
      "epoch": 2.411556603773585,
      "grad_norm": 1.5,
      "learning_rate": 0.00044178691255732707,
      "loss": 5.3424,
      "mean_token_accuracy": 0.17719968408346176,
      "num_tokens": 55157386.0,
      "step": 26585
    },
    {
      "entropy": 5.724086570739746,
      "epoch": 2.412010159651669,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004417651931631708,
      "loss": 5.4305,
      "mean_token_accuracy": 0.1732562467455864,
      "num_tokens": 55168969.0,
      "step": 26590
    },
    {
      "entropy": 5.671455240249633,
      "epoch": 2.4124637155297535,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004417434703202678,
      "loss": 5.3231,
      "mean_token_accuracy": 0.16925457417964934,
      "num_tokens": 55180123.0,
      "step": 26595
    },
    {
      "entropy": 5.66319432258606,
      "epoch": 2.4129172714078373,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004417217440290673,
      "loss": 5.2984,
      "mean_token_accuracy": 0.18163794726133348,
      "num_tokens": 55190668.0,
      "step": 26600
    },
    {
      "entropy": 5.6274330615997314,
      "epoch": 2.4133708272859216,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004417000142900188,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1846321165561676,
      "num_tokens": 55200852.0,
      "step": 26605
    },
    {
      "entropy": 5.738687419891358,
      "epoch": 2.413824383164006,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004416782811035715,
      "loss": 5.4483,
      "mean_token_accuracy": 0.17396497428417207,
      "num_tokens": 55211174.0,
      "step": 26610
    },
    {
      "entropy": 5.609450244903565,
      "epoch": 2.41427793904209,
      "grad_norm": 1.75,
      "learning_rate": 0.00044165654447017485,
      "loss": 5.2044,
      "mean_token_accuracy": 0.18517910987138747,
      "num_tokens": 55222042.0,
      "step": 26615
    },
    {
      "entropy": 5.678867673873901,
      "epoch": 2.4147314949201744,
      "grad_norm": 2.25,
      "learning_rate": 0.00044163480439027845,
      "loss": 5.3694,
      "mean_token_accuracy": 0.17515862435102464,
      "num_tokens": 55231688.0,
      "step": 26620
    },
    {
      "entropy": 5.701185321807861,
      "epoch": 2.415185050798258,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004416130608643317,
      "loss": 5.3254,
      "mean_token_accuracy": 0.18214766681194305,
      "num_tokens": 55241089.0,
      "step": 26625
    },
    {
      "entropy": 5.678887033462525,
      "epoch": 2.4156386066763424,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004415913138927844,
      "loss": 5.3197,
      "mean_token_accuracy": 0.18291959315538406,
      "num_tokens": 55251511.0,
      "step": 26630
    },
    {
      "entropy": 5.599879407882691,
      "epoch": 2.4160921625544267,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004415695634760864,
      "loss": 5.3132,
      "mean_token_accuracy": 0.17609254717826844,
      "num_tokens": 55262855.0,
      "step": 26635
    },
    {
      "entropy": 5.6101216793060305,
      "epoch": 2.416545718432511,
      "grad_norm": 2.25,
      "learning_rate": 0.0004415478096146873,
      "loss": 5.215,
      "mean_token_accuracy": 0.1872733101248741,
      "num_tokens": 55273035.0,
      "step": 26640
    },
    {
      "entropy": 5.634017658233643,
      "epoch": 2.4169992743105952,
      "grad_norm": 1.546875,
      "learning_rate": 0.000441526052309037,
      "loss": 5.3234,
      "mean_token_accuracy": 0.18325188905000686,
      "num_tokens": 55284033.0,
      "step": 26645
    },
    {
      "entropy": 5.675400590896606,
      "epoch": 2.417452830188679,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004415042915595856,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1864340841770172,
      "num_tokens": 55294290.0,
      "step": 26650
    },
    {
      "entropy": 5.694205093383789,
      "epoch": 2.4179063860667633,
      "grad_norm": 4.53125,
      "learning_rate": 0.000441482527366783,
      "loss": 5.3583,
      "mean_token_accuracy": 0.1779148131608963,
      "num_tokens": 55305681.0,
      "step": 26655
    },
    {
      "entropy": 5.58279938697815,
      "epoch": 2.4183599419448476,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004414607597310793,
      "loss": 5.3049,
      "mean_token_accuracy": 0.18167287856340408,
      "num_tokens": 55316647.0,
      "step": 26660
    },
    {
      "entropy": 5.731395578384399,
      "epoch": 2.418813497822932,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004414389886529247,
      "loss": 5.4542,
      "mean_token_accuracy": 0.17024235427379608,
      "num_tokens": 55328318.0,
      "step": 26665
    },
    {
      "entropy": 5.7306492805480955,
      "epoch": 2.419267053701016,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004414172141327693,
      "loss": 5.3505,
      "mean_token_accuracy": 0.16972139477729797,
      "num_tokens": 55338483.0,
      "step": 26670
    },
    {
      "entropy": 5.661090612411499,
      "epoch": 2.4197206095791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004413954361710636,
      "loss": 5.3907,
      "mean_token_accuracy": 0.17730941772460937,
      "num_tokens": 55348648.0,
      "step": 26675
    },
    {
      "entropy": 5.63946795463562,
      "epoch": 2.4201741654571842,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044137365476825785,
      "loss": 5.2709,
      "mean_token_accuracy": 0.1810607209801674,
      "num_tokens": 55358599.0,
      "step": 26680
    },
    {
      "entropy": 5.687184762954712,
      "epoch": 2.4206277213352685,
      "grad_norm": 1.75,
      "learning_rate": 0.00044135186992480253,
      "loss": 5.3734,
      "mean_token_accuracy": 0.1763969197869301,
      "num_tokens": 55369847.0,
      "step": 26685
    },
    {
      "entropy": 5.5982777118682865,
      "epoch": 2.4210812772133528,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044133008164114814,
      "loss": 5.3366,
      "mean_token_accuracy": 0.17860252559185028,
      "num_tokens": 55381235.0,
      "step": 26690
    },
    {
      "entropy": 5.680571746826172,
      "epoch": 2.421534833091437,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044130828991774526,
      "loss": 5.3481,
      "mean_token_accuracy": 0.1754855692386627,
      "num_tokens": 55391527.0,
      "step": 26695
    },
    {
      "entropy": 5.738258218765258,
      "epoch": 2.4219883889695213,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00044128649475504454,
      "loss": 5.4002,
      "mean_token_accuracy": 0.17648206353187562,
      "num_tokens": 55401528.0,
      "step": 26700
    },
    {
      "entropy": 5.740794277191162,
      "epoch": 2.422441944847605,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004412646961534968,
      "loss": 5.394,
      "mean_token_accuracy": 0.17585776150226592,
      "num_tokens": 55411411.0,
      "step": 26705
    },
    {
      "entropy": 5.5961329460144045,
      "epoch": 2.4228955007256894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004412428941135528,
      "loss": 5.1835,
      "mean_token_accuracy": 0.19071056693792343,
      "num_tokens": 55423164.0,
      "step": 26710
    },
    {
      "entropy": 5.651261234283448,
      "epoch": 2.4233490566037736,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004412210886356633,
      "loss": 5.2694,
      "mean_token_accuracy": 0.18871250748634338,
      "num_tokens": 55433397.0,
      "step": 26715
    },
    {
      "entropy": 5.600560426712036,
      "epoch": 2.423802612481858,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044119927972027936,
      "loss": 5.3013,
      "mean_token_accuracy": 0.1773083046078682,
      "num_tokens": 55442898.0,
      "step": 26720
    },
    {
      "entropy": 5.659531164169311,
      "epoch": 2.4242561683599417,
      "grad_norm": 2.296875,
      "learning_rate": 0.00044117746736785195,
      "loss": 5.3377,
      "mean_token_accuracy": 0.18141702115535735,
      "num_tokens": 55452226.0,
      "step": 26725
    },
    {
      "entropy": 5.714975833892822,
      "epoch": 2.424709724238026,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004411556515788322,
      "loss": 5.313,
      "mean_token_accuracy": 0.17427289336919785,
      "num_tokens": 55462503.0,
      "step": 26730
    },
    {
      "entropy": 5.7132669448852536,
      "epoch": 2.4251632801161103,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004411338323536712,
      "loss": 5.4758,
      "mean_token_accuracy": 0.1701494425535202,
      "num_tokens": 55472789.0,
      "step": 26735
    },
    {
      "entropy": 5.639616537094116,
      "epoch": 2.4256168359941945,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004411120096928203,
      "loss": 5.2672,
      "mean_token_accuracy": 0.18227705359458923,
      "num_tokens": 55482572.0,
      "step": 26740
    },
    {
      "entropy": 5.6766644477844235,
      "epoch": 2.426070391872279,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004410901835967306,
      "loss": 5.3252,
      "mean_token_accuracy": 0.18480826467275618,
      "num_tokens": 55493033.0,
      "step": 26745
    },
    {
      "entropy": 5.625788640975952,
      "epoch": 2.426523947750363,
      "grad_norm": 2.046875,
      "learning_rate": 0.00044106835406585354,
      "loss": 5.2918,
      "mean_token_accuracy": 0.1759827271103859,
      "num_tokens": 55503574.0,
      "step": 26750
    },
    {
      "entropy": 5.70268235206604,
      "epoch": 2.426977503628447,
      "grad_norm": 2.125,
      "learning_rate": 0.00044104652110064067,
      "loss": 5.3604,
      "mean_token_accuracy": 0.17349779605865479,
      "num_tokens": 55514119.0,
      "step": 26755
    },
    {
      "entropy": 5.638462686538697,
      "epoch": 2.427431059506531,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004410246847015435,
      "loss": 5.2779,
      "mean_token_accuracy": 0.17929334938526154,
      "num_tokens": 55524973.0,
      "step": 26760
    },
    {
      "entropy": 5.634920215606689,
      "epoch": 2.4278846153846154,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004410028448690134,
      "loss": 5.3828,
      "mean_token_accuracy": 0.17755352705717087,
      "num_tokens": 55534847.0,
      "step": 26765
    },
    {
      "entropy": 5.695801067352295,
      "epoch": 2.4283381712626997,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004409810016035023,
      "loss": 5.3462,
      "mean_token_accuracy": 0.173136343061924,
      "num_tokens": 55544886.0,
      "step": 26770
    },
    {
      "entropy": 5.670726680755616,
      "epoch": 2.4287917271407835,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004409591549054617,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1790713518857956,
      "num_tokens": 55554534.0,
      "step": 26775
    },
    {
      "entropy": 5.722982120513916,
      "epoch": 2.4292452830188678,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044093730477534355,
      "loss": 5.3453,
      "mean_token_accuracy": 0.17492507994174958,
      "num_tokens": 55564302.0,
      "step": 26780
    },
    {
      "entropy": 5.616900682449341,
      "epoch": 2.429698838896952,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004409154512135996,
      "loss": 5.264,
      "mean_token_accuracy": 0.1813106805086136,
      "num_tokens": 55574543.0,
      "step": 26785
    },
    {
      "entropy": 5.549363994598389,
      "epoch": 2.4301523947750363,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004408935942206818,
      "loss": 5.2767,
      "mean_token_accuracy": 0.1830565258860588,
      "num_tokens": 55583849.0,
      "step": 26790
    },
    {
      "entropy": 5.658525037765503,
      "epoch": 2.4306059506531206,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004408717337970423,
      "loss": 5.3536,
      "mean_token_accuracy": 0.17875595688819884,
      "num_tokens": 55594294.0,
      "step": 26795
    },
    {
      "entropy": 5.701722908020019,
      "epoch": 2.431059506531205,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044084986994313304,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1760143220424652,
      "num_tokens": 55604185.0,
      "step": 26800
    },
    {
      "entropy": 5.664311361312866,
      "epoch": 2.4315130624092887,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004408280026594062,
      "loss": 5.2802,
      "mean_token_accuracy": 0.18076078593730927,
      "num_tokens": 55615203.0,
      "step": 26805
    },
    {
      "entropy": 5.585596227645874,
      "epoch": 2.431966618287373,
      "grad_norm": 2.03125,
      "learning_rate": 0.000440806131946314,
      "loss": 5.3934,
      "mean_token_accuracy": 0.1754080221056938,
      "num_tokens": 55626157.0,
      "step": 26810
    },
    {
      "entropy": 5.734741449356079,
      "epoch": 2.432420174165457,
      "grad_norm": 1.9375,
      "learning_rate": 0.00044078425780430877,
      "loss": 5.3461,
      "mean_token_accuracy": 0.1744267538189888,
      "num_tokens": 55636862.0,
      "step": 26815
    },
    {
      "entropy": 5.668906354904175,
      "epoch": 2.4328737300435415,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004407623802338428,
      "loss": 5.2625,
      "mean_token_accuracy": 0.18176039457321166,
      "num_tokens": 55647646.0,
      "step": 26820
    },
    {
      "entropy": 5.677312803268433,
      "epoch": 2.4333272859216257,
      "grad_norm": 2.375,
      "learning_rate": 0.00044074049923536855,
      "loss": 5.3294,
      "mean_token_accuracy": 0.1823972910642624,
      "num_tokens": 55659847.0,
      "step": 26825
    },
    {
      "entropy": 5.643702268600464,
      "epoch": 2.4337808417997095,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004407186148093386,
      "loss": 5.3601,
      "mean_token_accuracy": 0.17712535262107848,
      "num_tokens": 55670626.0,
      "step": 26830
    },
    {
      "entropy": 5.665673303604126,
      "epoch": 2.434234397677794,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044069672695620543,
      "loss": 5.3926,
      "mean_token_accuracy": 0.1772778868675232,
      "num_tokens": 55681368.0,
      "step": 26835
    },
    {
      "entropy": 5.642350673675537,
      "epoch": 2.434687953555878,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004406748356764217,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18029594868421556,
      "num_tokens": 55691606.0,
      "step": 26840
    },
    {
      "entropy": 5.683452796936035,
      "epoch": 2.4351415094339623,
      "grad_norm": 2.375,
      "learning_rate": 0.00044065294097044017,
      "loss": 5.4175,
      "mean_token_accuracy": 0.17844967246055604,
      "num_tokens": 55702223.0,
      "step": 26845
    },
    {
      "entropy": 5.663005256652832,
      "epoch": 2.4355950653120466,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004406310428387136,
      "loss": 5.243,
      "mean_token_accuracy": 0.18000574111938478,
      "num_tokens": 55712729.0,
      "step": 26850
    },
    {
      "entropy": 5.613226652145386,
      "epoch": 2.4360486211901304,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004406091412816949,
      "loss": 5.333,
      "mean_token_accuracy": 0.1826530024409294,
      "num_tokens": 55722863.0,
      "step": 26855
    },
    {
      "entropy": 5.709610843658448,
      "epoch": 2.4365021770682147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044058723629983694,
      "loss": 5.4215,
      "mean_token_accuracy": 0.18291306644678115,
      "num_tokens": 55733124.0,
      "step": 26860
    },
    {
      "entropy": 5.761013746261597,
      "epoch": 2.436955732946299,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004405653278935927,
      "loss": 5.346,
      "mean_token_accuracy": 0.18496277034282685,
      "num_tokens": 55743438.0,
      "step": 26865
    },
    {
      "entropy": 5.6080522537231445,
      "epoch": 2.4374092888243832,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044054341606341533,
      "loss": 5.2067,
      "mean_token_accuracy": 0.181036975979805,
      "num_tokens": 55753602.0,
      "step": 26870
    },
    {
      "entropy": 5.636270952224732,
      "epoch": 2.4378628447024675,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004405215008097579,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18515910506248473,
      "num_tokens": 55763473.0,
      "step": 26875
    },
    {
      "entropy": 5.656379556655883,
      "epoch": 2.4383164005805513,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044049958213307364,
      "loss": 5.2842,
      "mean_token_accuracy": 0.17740925699472426,
      "num_tokens": 55772848.0,
      "step": 26880
    },
    {
      "entropy": 5.5733644485473635,
      "epoch": 2.4387699564586356,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044047766003381584,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18198008239269256,
      "num_tokens": 55783468.0,
      "step": 26885
    },
    {
      "entropy": 5.645589303970337,
      "epoch": 2.43922351233672,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00044045573451243793,
      "loss": 5.3449,
      "mean_token_accuracy": 0.18582651913166046,
      "num_tokens": 55793619.0,
      "step": 26890
    },
    {
      "entropy": 5.710954427719116,
      "epoch": 2.439677068214804,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044043380556939326,
      "loss": 5.3161,
      "mean_token_accuracy": 0.17933154106140137,
      "num_tokens": 55803869.0,
      "step": 26895
    },
    {
      "entropy": 5.717155599594117,
      "epoch": 2.4401306240928884,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004404118732051353,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17986734360456466,
      "num_tokens": 55814754.0,
      "step": 26900
    },
    {
      "entropy": 5.622897434234619,
      "epoch": 2.4405841799709727,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004403899374201176,
      "loss": 5.2851,
      "mean_token_accuracy": 0.17675640881061555,
      "num_tokens": 55824862.0,
      "step": 26905
    },
    {
      "entropy": 5.549218988418579,
      "epoch": 2.4410377358490565,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044036799821479397,
      "loss": 5.2762,
      "mean_token_accuracy": 0.17872394025325775,
      "num_tokens": 55835506.0,
      "step": 26910
    },
    {
      "entropy": 5.699302625656128,
      "epoch": 2.4414912917271407,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044034605558961786,
      "loss": 5.3978,
      "mean_token_accuracy": 0.17419212013483049,
      "num_tokens": 55846244.0,
      "step": 26915
    },
    {
      "entropy": 5.6594442367553714,
      "epoch": 2.441944847605225,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044032410954504325,
      "loss": 5.2727,
      "mean_token_accuracy": 0.1837437316775322,
      "num_tokens": 55856192.0,
      "step": 26920
    },
    {
      "entropy": 5.653914737701416,
      "epoch": 2.4423984034833093,
      "grad_norm": 3.34375,
      "learning_rate": 0.00044030216008152394,
      "loss": 5.3233,
      "mean_token_accuracy": 0.17996551245450973,
      "num_tokens": 55866509.0,
      "step": 26925
    },
    {
      "entropy": 5.632760620117187,
      "epoch": 2.442851959361393,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044028020719951376,
      "loss": 5.3415,
      "mean_token_accuracy": 0.179194812476635,
      "num_tokens": 55877321.0,
      "step": 26930
    },
    {
      "entropy": 5.659947109222412,
      "epoch": 2.4433055152394774,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004402582508994669,
      "loss": 5.3283,
      "mean_token_accuracy": 0.1779287725687027,
      "num_tokens": 55887933.0,
      "step": 26935
    },
    {
      "entropy": 5.628340816497802,
      "epoch": 2.4437590711175616,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044023629118183726,
      "loss": 5.2681,
      "mean_token_accuracy": 0.17722510695457458,
      "num_tokens": 55897274.0,
      "step": 26940
    },
    {
      "entropy": 5.671986818313599,
      "epoch": 2.444212626995646,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044021432804707895,
      "loss": 5.4088,
      "mean_token_accuracy": 0.1725396618247032,
      "num_tokens": 55907393.0,
      "step": 26945
    },
    {
      "entropy": 5.623407888412475,
      "epoch": 2.44466618287373,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004401923614956463,
      "loss": 5.317,
      "mean_token_accuracy": 0.17690256536006926,
      "num_tokens": 55918429.0,
      "step": 26950
    },
    {
      "entropy": 5.714226722717285,
      "epoch": 2.4451197387518144,
      "grad_norm": 1.625,
      "learning_rate": 0.0004401703915279935,
      "loss": 5.2793,
      "mean_token_accuracy": 0.1804795131087303,
      "num_tokens": 55928733.0,
      "step": 26955
    },
    {
      "entropy": 5.654043102264405,
      "epoch": 2.4455732946298983,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044014841814457486,
      "loss": 5.2701,
      "mean_token_accuracy": 0.18496992588043212,
      "num_tokens": 55939128.0,
      "step": 26960
    },
    {
      "entropy": 5.682537698745728,
      "epoch": 2.4460268505079825,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004401264413458449,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1782919555902481,
      "num_tokens": 55950237.0,
      "step": 26965
    },
    {
      "entropy": 5.698459768295288,
      "epoch": 2.446480406386067,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004401044611322581,
      "loss": 5.2508,
      "mean_token_accuracy": 0.1836119681596756,
      "num_tokens": 55960436.0,
      "step": 26970
    },
    {
      "entropy": 5.6393648147583,
      "epoch": 2.446933962264151,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004400824775042689,
      "loss": 5.2256,
      "mean_token_accuracy": 0.18273597359657287,
      "num_tokens": 55970446.0,
      "step": 26975
    },
    {
      "entropy": 5.65643424987793,
      "epoch": 2.447387518142235,
      "grad_norm": 2.5,
      "learning_rate": 0.000440060490462332,
      "loss": 5.343,
      "mean_token_accuracy": 0.1798518031835556,
      "num_tokens": 55980706.0,
      "step": 26980
    },
    {
      "entropy": 5.701181888580322,
      "epoch": 2.447841074020319,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004400385000069021,
      "loss": 5.3813,
      "mean_token_accuracy": 0.17297212332487105,
      "num_tokens": 55990669.0,
      "step": 26985
    },
    {
      "entropy": 5.740975999832154,
      "epoch": 2.4482946298984034,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044001650613843395,
      "loss": 5.3978,
      "mean_token_accuracy": 0.1756839171051979,
      "num_tokens": 56000709.0,
      "step": 26990
    },
    {
      "entropy": 5.656298732757568,
      "epoch": 2.4487481857764877,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004399945088573824,
      "loss": 5.3176,
      "mean_token_accuracy": 0.1802394926548004,
      "num_tokens": 56011435.0,
      "step": 26995
    },
    {
      "entropy": 5.688565063476562,
      "epoch": 2.449201741654572,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043997250816420246,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1854260340332985,
      "num_tokens": 56021264.0,
      "step": 27000
    },
    {
      "epoch": 2.449201741654572,
      "eval_entropy": 5.455870797299691,
      "eval_loss": 5.39945125579834,
      "eval_mean_token_accuracy": 0.18448574134144444,
      "eval_num_tokens": 56021264.0,
      "eval_runtime": 26.6129,
      "eval_samples_per_second": 1328.68,
      "eval_steps_per_second": 166.085,
      "step": 27000
    },
    {
      "entropy": 5.647222232818604,
      "epoch": 2.449655297532656,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004399505040593489,
      "loss": 5.3443,
      "mean_token_accuracy": 0.17911190539598465,
      "num_tokens": 56031463.0,
      "step": 27005
    },
    {
      "entropy": 5.597201871871948,
      "epoch": 2.45010885341074,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00043992849654327695,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18634404093027115,
      "num_tokens": 56041195.0,
      "step": 27010
    },
    {
      "entropy": 5.635239553451538,
      "epoch": 2.4505624092888243,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004399064856164416,
      "loss": 5.3433,
      "mean_token_accuracy": 0.17738768607378005,
      "num_tokens": 56051840.0,
      "step": 27015
    },
    {
      "entropy": 5.700620365142822,
      "epoch": 2.4510159651669086,
      "grad_norm": 1.78125,
      "learning_rate": 0.00043988447127929817,
      "loss": 5.366,
      "mean_token_accuracy": 0.17514799982309343,
      "num_tokens": 56062535.0,
      "step": 27020
    },
    {
      "entropy": 5.675134325027466,
      "epoch": 2.451469521044993,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043986245353230173,
      "loss": 5.2823,
      "mean_token_accuracy": 0.18881472051143647,
      "num_tokens": 56072177.0,
      "step": 27025
    },
    {
      "entropy": 5.561778402328491,
      "epoch": 2.451923076923077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043984043237590785,
      "loss": 5.3307,
      "mean_token_accuracy": 0.17971904426813126,
      "num_tokens": 56081607.0,
      "step": 27030
    },
    {
      "entropy": 5.6556782722473145,
      "epoch": 2.452376632801161,
      "grad_norm": 1.625,
      "learning_rate": 0.00043981840781057185,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1776565507054329,
      "num_tokens": 56093371.0,
      "step": 27035
    },
    {
      "entropy": 5.674274921417236,
      "epoch": 2.452830188679245,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043979637983674913,
      "loss": 5.3177,
      "mean_token_accuracy": 0.18005963563919067,
      "num_tokens": 56103973.0,
      "step": 27040
    },
    {
      "entropy": 5.661172485351562,
      "epoch": 2.4532837445573294,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004397743484548953,
      "loss": 5.3175,
      "mean_token_accuracy": 0.17643149942159653,
      "num_tokens": 56113906.0,
      "step": 27045
    },
    {
      "entropy": 5.6571540355682375,
      "epoch": 2.4537373004354137,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004397523136654659,
      "loss": 5.2995,
      "mean_token_accuracy": 0.17930044680833818,
      "num_tokens": 56124172.0,
      "step": 27050
    },
    {
      "entropy": 5.598284196853638,
      "epoch": 2.454190856313498,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043973027546891677,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19316072762012482,
      "num_tokens": 56133088.0,
      "step": 27055
    },
    {
      "entropy": 5.556800365447998,
      "epoch": 2.4546444121915822,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004397082338657034,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18822112679481506,
      "num_tokens": 56143098.0,
      "step": 27060
    },
    {
      "entropy": 5.670928621292115,
      "epoch": 2.455097968069666,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00043968618885628185,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1816401168704033,
      "num_tokens": 56153889.0,
      "step": 27065
    },
    {
      "entropy": 5.591073513031006,
      "epoch": 2.4555515239477503,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043966414044110805,
      "loss": 5.2684,
      "mean_token_accuracy": 0.1851788267493248,
      "num_tokens": 56164164.0,
      "step": 27070
    },
    {
      "entropy": 5.665050077438354,
      "epoch": 2.4560050798258346,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043964208862063775,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17917057275772094,
      "num_tokens": 56173172.0,
      "step": 27075
    },
    {
      "entropy": 5.6497049808502195,
      "epoch": 2.456458635703919,
      "grad_norm": 2.03125,
      "learning_rate": 0.00043962003339532714,
      "loss": 5.3184,
      "mean_token_accuracy": 0.18597880452871324,
      "num_tokens": 56183127.0,
      "step": 27080
    },
    {
      "entropy": 5.644375562667847,
      "epoch": 2.4569121915820027,
      "grad_norm": 2.25,
      "learning_rate": 0.00043959797476563226,
      "loss": 5.3413,
      "mean_token_accuracy": 0.1791798174381256,
      "num_tokens": 56192528.0,
      "step": 27085
    },
    {
      "entropy": 5.605853509902954,
      "epoch": 2.457365747460087,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00043957591273200943,
      "loss": 5.2963,
      "mean_token_accuracy": 0.17935230880975722,
      "num_tokens": 56203239.0,
      "step": 27090
    },
    {
      "entropy": 5.712042617797851,
      "epoch": 2.4578193033381712,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004395538472949146,
      "loss": 5.3607,
      "mean_token_accuracy": 0.17250601798295975,
      "num_tokens": 56214068.0,
      "step": 27095
    },
    {
      "entropy": 5.737274551391602,
      "epoch": 2.4582728592162555,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004395317784548045,
      "loss": 5.3325,
      "mean_token_accuracy": 0.17189761102199555,
      "num_tokens": 56224829.0,
      "step": 27100
    },
    {
      "entropy": 5.665706825256348,
      "epoch": 2.4587264150943398,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004395097062121351,
      "loss": 5.3539,
      "mean_token_accuracy": 0.18075544983148575,
      "num_tokens": 56235412.0,
      "step": 27105
    },
    {
      "entropy": 5.708569574356079,
      "epoch": 2.459179970972424,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043948763056736315,
      "loss": 5.3907,
      "mean_token_accuracy": 0.17468624860048293,
      "num_tokens": 56246785.0,
      "step": 27110
    },
    {
      "entropy": 5.680214262008667,
      "epoch": 2.459633526850508,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000439465551520945,
      "loss": 5.3716,
      "mean_token_accuracy": 0.17550163865089416,
      "num_tokens": 56256701.0,
      "step": 27115
    },
    {
      "entropy": 5.652216863632202,
      "epoch": 2.460087082728592,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043944346907333744,
      "loss": 5.3196,
      "mean_token_accuracy": 0.18341683000326156,
      "num_tokens": 56267418.0,
      "step": 27120
    },
    {
      "entropy": 5.591608905792237,
      "epoch": 2.4605406386066764,
      "grad_norm": 1.578125,
      "learning_rate": 0.000439421383224997,
      "loss": 5.2451,
      "mean_token_accuracy": 0.1834714338183403,
      "num_tokens": 56278232.0,
      "step": 27125
    },
    {
      "entropy": 5.6978973865509035,
      "epoch": 2.4609941944847606,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004393992939763804,
      "loss": 5.3299,
      "mean_token_accuracy": 0.1856088548898697,
      "num_tokens": 56289866.0,
      "step": 27130
    },
    {
      "entropy": 5.691002655029297,
      "epoch": 2.4614477503628445,
      "grad_norm": 2.125,
      "learning_rate": 0.00043937720132794457,
      "loss": 5.2813,
      "mean_token_accuracy": 0.17948664873838424,
      "num_tokens": 56299595.0,
      "step": 27135
    },
    {
      "entropy": 5.72111668586731,
      "epoch": 2.4619013062409287,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043935510528014625,
      "loss": 5.3844,
      "mean_token_accuracy": 0.18043726235628127,
      "num_tokens": 56310152.0,
      "step": 27140
    },
    {
      "entropy": 5.664232921600342,
      "epoch": 2.462354862119013,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004393330058334426,
      "loss": 5.4016,
      "mean_token_accuracy": 0.17173661589622496,
      "num_tokens": 56320288.0,
      "step": 27145
    },
    {
      "entropy": 5.7145960330963135,
      "epoch": 2.4628084179970973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043931090298829047,
      "loss": 5.3457,
      "mean_token_accuracy": 0.17905534207820892,
      "num_tokens": 56330345.0,
      "step": 27150
    },
    {
      "entropy": 5.719800806045532,
      "epoch": 2.4632619738751815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043928879674514694,
      "loss": 5.4547,
      "mean_token_accuracy": 0.18278600573539733,
      "num_tokens": 56340996.0,
      "step": 27155
    },
    {
      "entropy": 5.502612113952637,
      "epoch": 2.463715529753266,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004392666871044692,
      "loss": 5.1881,
      "mean_token_accuracy": 0.19088293612003326,
      "num_tokens": 56350326.0,
      "step": 27160
    },
    {
      "entropy": 5.663477563858033,
      "epoch": 2.4641690856313496,
      "grad_norm": 2.125,
      "learning_rate": 0.00043924457406671456,
      "loss": 5.3387,
      "mean_token_accuracy": 0.18320660442113876,
      "num_tokens": 56360556.0,
      "step": 27165
    },
    {
      "entropy": 5.6773487567901615,
      "epoch": 2.464622641509434,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004392224576323403,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1826433077454567,
      "num_tokens": 56369980.0,
      "step": 27170
    },
    {
      "entropy": 5.626096248626709,
      "epoch": 2.465076197387518,
      "grad_norm": 1.90625,
      "learning_rate": 0.00043920033780180367,
      "loss": 5.3405,
      "mean_token_accuracy": 0.17399940639734268,
      "num_tokens": 56379974.0,
      "step": 27175
    },
    {
      "entropy": 5.640738487243652,
      "epoch": 2.4655297532656024,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004391782145755623,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18428014665842057,
      "num_tokens": 56390542.0,
      "step": 27180
    },
    {
      "entropy": 5.736296796798706,
      "epoch": 2.4659833091436867,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004391560879540736,
      "loss": 5.4422,
      "mean_token_accuracy": 0.17038571685552598,
      "num_tokens": 56401755.0,
      "step": 27185
    },
    {
      "entropy": 5.64190526008606,
      "epoch": 2.4664368650217705,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004391339579377952,
      "loss": 5.3226,
      "mean_token_accuracy": 0.17968318462371827,
      "num_tokens": 56412223.0,
      "step": 27190
    },
    {
      "entropy": 5.680315113067627,
      "epoch": 2.4668904208998548,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043911182452718476,
      "loss": 5.3524,
      "mean_token_accuracy": 0.1800047814846039,
      "num_tokens": 56422947.0,
      "step": 27195
    },
    {
      "entropy": 5.72771053314209,
      "epoch": 2.467343976777939,
      "grad_norm": 1.90625,
      "learning_rate": 0.00043908968772269984,
      "loss": 5.3409,
      "mean_token_accuracy": 0.184209206700325,
      "num_tokens": 56433189.0,
      "step": 27200
    },
    {
      "entropy": 5.682176160812378,
      "epoch": 2.4677975326560233,
      "grad_norm": 2.265625,
      "learning_rate": 0.00043906754752479836,
      "loss": 5.3385,
      "mean_token_accuracy": 0.1858372762799263,
      "num_tokens": 56444460.0,
      "step": 27205
    },
    {
      "entropy": 5.649408674240112,
      "epoch": 2.4682510885341076,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004390454039339383,
      "loss": 5.3593,
      "mean_token_accuracy": 0.1770341157913208,
      "num_tokens": 56454368.0,
      "step": 27210
    },
    {
      "entropy": 5.59519591331482,
      "epoch": 2.4687046444121914,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00043902325695057743,
      "loss": 5.266,
      "mean_token_accuracy": 0.18263538926839828,
      "num_tokens": 56464153.0,
      "step": 27215
    },
    {
      "entropy": 5.634449815750122,
      "epoch": 2.4691582002902757,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043900110657517377,
      "loss": 5.2779,
      "mean_token_accuracy": 0.18583327382802964,
      "num_tokens": 56473627.0,
      "step": 27220
    },
    {
      "entropy": 5.595127058029175,
      "epoch": 2.46961175616836,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004389789528081855,
      "loss": 5.2793,
      "mean_token_accuracy": 0.18031456023454667,
      "num_tokens": 56483660.0,
      "step": 27225
    },
    {
      "entropy": 5.6466044902801515,
      "epoch": 2.470065312046444,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004389567956500707,
      "loss": 5.2505,
      "mean_token_accuracy": 0.1883341208100319,
      "num_tokens": 56492653.0,
      "step": 27230
    },
    {
      "entropy": 5.7676502704620365,
      "epoch": 2.4705188679245285,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004389346351012876,
      "loss": 5.4501,
      "mean_token_accuracy": 0.16902021765708924,
      "num_tokens": 56503289.0,
      "step": 27235
    },
    {
      "entropy": 5.676271057128906,
      "epoch": 2.4709724238026123,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00043891247116229436,
      "loss": 5.3964,
      "mean_token_accuracy": 0.16960871070623398,
      "num_tokens": 56513843.0,
      "step": 27240
    },
    {
      "entropy": 5.70573935508728,
      "epoch": 2.4714259796806965,
      "grad_norm": 2.15625,
      "learning_rate": 0.00043889030383354957,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17525189071893693,
      "num_tokens": 56524175.0,
      "step": 27245
    },
    {
      "entropy": 5.667827796936035,
      "epoch": 2.471879535558781,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004388681331155114,
      "loss": 5.2702,
      "mean_token_accuracy": 0.18840693980455397,
      "num_tokens": 56534976.0,
      "step": 27250
    },
    {
      "entropy": 5.697813892364502,
      "epoch": 2.472333091436865,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004388459590086386,
      "loss": 5.3695,
      "mean_token_accuracy": 0.1762072190642357,
      "num_tokens": 56545967.0,
      "step": 27255
    },
    {
      "entropy": 5.654846954345703,
      "epoch": 2.4727866473149493,
      "grad_norm": 1.625,
      "learning_rate": 0.00043882378151338956,
      "loss": 5.3324,
      "mean_token_accuracy": 0.18012223690748214,
      "num_tokens": 56555527.0,
      "step": 27260
    },
    {
      "entropy": 5.671675682067871,
      "epoch": 2.4732402031930336,
      "grad_norm": 2.5625,
      "learning_rate": 0.000438801600630223,
      "loss": 5.3469,
      "mean_token_accuracy": 0.1714993193745613,
      "num_tokens": 56565874.0,
      "step": 27265
    },
    {
      "entropy": 5.614894676208496,
      "epoch": 2.4736937590711174,
      "grad_norm": 1.734375,
      "learning_rate": 0.00043877941635959755,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18245447874069215,
      "num_tokens": 56575452.0,
      "step": 27270
    },
    {
      "entropy": 5.602878856658935,
      "epoch": 2.4741473149492017,
      "grad_norm": 1.90625,
      "learning_rate": 0.00043875722870197214,
      "loss": 5.2548,
      "mean_token_accuracy": 0.18323417156934738,
      "num_tokens": 56586176.0,
      "step": 27275
    },
    {
      "entropy": 5.683250427246094,
      "epoch": 2.474600870827286,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004387350376578054,
      "loss": 5.3016,
      "mean_token_accuracy": 0.1804349437355995,
      "num_tokens": 56595473.0,
      "step": 27280
    },
    {
      "entropy": 5.617850208282471,
      "epoch": 2.4750544267053702,
      "grad_norm": 4.875,
      "learning_rate": 0.0004387128432275563,
      "loss": 5.3609,
      "mean_token_accuracy": 0.17470623105764388,
      "num_tokens": 56606016.0,
      "step": 27285
    },
    {
      "entropy": 5.658481502532959,
      "epoch": 2.475507982583454,
      "grad_norm": 1.796875,
      "learning_rate": 0.000438690645411684,
      "loss": 5.352,
      "mean_token_accuracy": 0.1746321737766266,
      "num_tokens": 56617674.0,
      "step": 27290
    },
    {
      "entropy": 5.667084884643555,
      "epoch": 2.4759615384615383,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043866844421064745,
      "loss": 5.3006,
      "mean_token_accuracy": 0.18077594488859178,
      "num_tokens": 56628077.0,
      "step": 27295
    },
    {
      "entropy": 5.670559740066528,
      "epoch": 2.4764150943396226,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043864623962490575,
      "loss": 5.2743,
      "mean_token_accuracy": 0.18283104300498962,
      "num_tokens": 56638691.0,
      "step": 27300
    },
    {
      "entropy": 5.629169559478759,
      "epoch": 2.476868650217707,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004386240316549181,
      "loss": 5.3176,
      "mean_token_accuracy": 0.17857445478439332,
      "num_tokens": 56650338.0,
      "step": 27305
    },
    {
      "entropy": 5.627089166641236,
      "epoch": 2.477322206095791,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043860182030114374,
      "loss": 5.2518,
      "mean_token_accuracy": 0.18348638862371444,
      "num_tokens": 56660328.0,
      "step": 27310
    },
    {
      "entropy": 5.763763523101806,
      "epoch": 2.4777757619738754,
      "grad_norm": 4.1875,
      "learning_rate": 0.00043857960556404214,
      "loss": 5.394,
      "mean_token_accuracy": 0.183093886077404,
      "num_tokens": 56670674.0,
      "step": 27315
    },
    {
      "entropy": 5.658018779754639,
      "epoch": 2.478229317851959,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004385573874440725,
      "loss": 5.2813,
      "mean_token_accuracy": 0.18777859210968018,
      "num_tokens": 56680656.0,
      "step": 27320
    },
    {
      "entropy": 5.578556108474731,
      "epoch": 2.4786828737300435,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043853516594169446,
      "loss": 5.2443,
      "mean_token_accuracy": 0.18461072742938994,
      "num_tokens": 56691039.0,
      "step": 27325
    },
    {
      "entropy": 5.639869022369385,
      "epoch": 2.4791364296081277,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004385129410573675,
      "loss": 5.3351,
      "mean_token_accuracy": 0.17590806484222413,
      "num_tokens": 56703101.0,
      "step": 27330
    },
    {
      "entropy": 5.673494052886963,
      "epoch": 2.479589985486212,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043849071279155124,
      "loss": 5.3683,
      "mean_token_accuracy": 0.1840483471751213,
      "num_tokens": 56712351.0,
      "step": 27335
    },
    {
      "entropy": 5.640524291992188,
      "epoch": 2.480043541364296,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043846848114470537,
      "loss": 5.3396,
      "mean_token_accuracy": 0.1860524982213974,
      "num_tokens": 56722836.0,
      "step": 27340
    },
    {
      "entropy": 5.689134931564331,
      "epoch": 2.48049709724238,
      "grad_norm": 2.265625,
      "learning_rate": 0.00043844624611728974,
      "loss": 5.3198,
      "mean_token_accuracy": 0.1745445340871811,
      "num_tokens": 56734032.0,
      "step": 27345
    },
    {
      "entropy": 5.636423015594483,
      "epoch": 2.4809506531204644,
      "grad_norm": 1.9375,
      "learning_rate": 0.00043842400770976406,
      "loss": 5.2903,
      "mean_token_accuracy": 0.17934955954551696,
      "num_tokens": 56743536.0,
      "step": 27350
    },
    {
      "entropy": 5.633323860168457,
      "epoch": 2.4814042089985486,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004384017659225882,
      "loss": 5.3017,
      "mean_token_accuracy": 0.18197727650403978,
      "num_tokens": 56753475.0,
      "step": 27355
    },
    {
      "entropy": 5.5847372055053714,
      "epoch": 2.481857764876633,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043837952075622224,
      "loss": 5.2533,
      "mean_token_accuracy": 0.18292408734560012,
      "num_tokens": 56763602.0,
      "step": 27360
    },
    {
      "entropy": 5.691777229309082,
      "epoch": 2.482311320754717,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004383572722111261,
      "loss": 5.2555,
      "mean_token_accuracy": 0.18623434454202653,
      "num_tokens": 56773859.0,
      "step": 27365
    },
    {
      "entropy": 5.654943943023682,
      "epoch": 2.482764876632801,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043833502028775995,
      "loss": 5.3073,
      "mean_token_accuracy": 0.18350669890642166,
      "num_tokens": 56783810.0,
      "step": 27370
    },
    {
      "entropy": 5.665414810180664,
      "epoch": 2.4832184325108853,
      "grad_norm": 2.109375,
      "learning_rate": 0.00043831276498658395,
      "loss": 5.3827,
      "mean_token_accuracy": 0.17266375720500945,
      "num_tokens": 56794088.0,
      "step": 27375
    },
    {
      "entropy": 5.616735935211182,
      "epoch": 2.4836719883889695,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004382905063080584,
      "loss": 5.248,
      "mean_token_accuracy": 0.18526985943317414,
      "num_tokens": 56804623.0,
      "step": 27380
    },
    {
      "entropy": 5.606601715087891,
      "epoch": 2.484125544267054,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00043826824425264357,
      "loss": 5.3158,
      "mean_token_accuracy": 0.17609558254480362,
      "num_tokens": 56813842.0,
      "step": 27385
    },
    {
      "entropy": 5.682540559768677,
      "epoch": 2.484579100145138,
      "grad_norm": 2.5,
      "learning_rate": 0.0004382459788207997,
      "loss": 5.3116,
      "mean_token_accuracy": 0.1809016600251198,
      "num_tokens": 56823501.0,
      "step": 27390
    },
    {
      "entropy": 5.657256603240967,
      "epoch": 2.485032656023222,
      "grad_norm": 2.265625,
      "learning_rate": 0.00043822371001298754,
      "loss": 5.3302,
      "mean_token_accuracy": 0.1845744416117668,
      "num_tokens": 56834601.0,
      "step": 27395
    },
    {
      "entropy": 5.597074699401856,
      "epoch": 2.485486211901306,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004382014378296674,
      "loss": 5.3231,
      "mean_token_accuracy": 0.1784182921051979,
      "num_tokens": 56844701.0,
      "step": 27400
    },
    {
      "entropy": 5.5711112976074215,
      "epoch": 2.4859397677793904,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004381791622713,
      "loss": 5.2889,
      "mean_token_accuracy": 0.18453994244337082,
      "num_tokens": 56854867.0,
      "step": 27405
    },
    {
      "entropy": 5.728708648681641,
      "epoch": 2.4863933236574747,
      "grad_norm": 1.90625,
      "learning_rate": 0.000438156883338346,
      "loss": 5.3527,
      "mean_token_accuracy": 0.1812087446451187,
      "num_tokens": 56865688.0,
      "step": 27410
    },
    {
      "entropy": 5.7116083145141605,
      "epoch": 2.486846879535559,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043813460103126595,
      "loss": 5.3525,
      "mean_token_accuracy": 0.17845408171415328,
      "num_tokens": 56876504.0,
      "step": 27415
    },
    {
      "entropy": 5.6234746932983395,
      "epoch": 2.487300435413643,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043811231535052076,
      "loss": 5.2768,
      "mean_token_accuracy": 0.1831117480993271,
      "num_tokens": 56886083.0,
      "step": 27420
    },
    {
      "entropy": 5.670764636993408,
      "epoch": 2.487753991291727,
      "grad_norm": 3.0,
      "learning_rate": 0.0004380900262965714,
      "loss": 5.3897,
      "mean_token_accuracy": 0.17800740003585816,
      "num_tokens": 56897079.0,
      "step": 27425
    },
    {
      "entropy": 5.637181663513184,
      "epoch": 2.4882075471698113,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004380677338698787,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1845424920320511,
      "num_tokens": 56906962.0,
      "step": 27430
    },
    {
      "entropy": 5.688675403594971,
      "epoch": 2.4886611030478956,
      "grad_norm": 1.953125,
      "learning_rate": 0.00043804543807090374,
      "loss": 5.4104,
      "mean_token_accuracy": 0.17729368805885315,
      "num_tokens": 56918156.0,
      "step": 27435
    },
    {
      "entropy": 5.632278776168823,
      "epoch": 2.48911465892598,
      "grad_norm": 2.953125,
      "learning_rate": 0.00043802313890010757,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17903127670288085,
      "num_tokens": 56928888.0,
      "step": 27440
    },
    {
      "entropy": 5.66468358039856,
      "epoch": 2.4895682148040637,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004380008363579513,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17782489210367203,
      "num_tokens": 56939929.0,
      "step": 27445
    },
    {
      "entropy": 5.67213282585144,
      "epoch": 2.490021770682148,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043797853044489623,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18238457441329955,
      "num_tokens": 56950992.0,
      "step": 27450
    },
    {
      "entropy": 5.676809787750244,
      "epoch": 2.490475326560232,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004379562211614036,
      "loss": 5.4227,
      "mean_token_accuracy": 0.17995397597551346,
      "num_tokens": 56962118.0,
      "step": 27455
    },
    {
      "entropy": 5.664448499679565,
      "epoch": 2.4909288824383164,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004379339085079348,
      "loss": 5.3028,
      "mean_token_accuracy": 0.18720235377550126,
      "num_tokens": 56972837.0,
      "step": 27460
    },
    {
      "entropy": 5.680055570602417,
      "epoch": 2.4913824383164007,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004379115924849512,
      "loss": 5.3071,
      "mean_token_accuracy": 0.18104926645755767,
      "num_tokens": 56982794.0,
      "step": 27465
    },
    {
      "entropy": 5.76543607711792,
      "epoch": 2.491835994194485,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004378892730929145,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1768724128603935,
      "num_tokens": 56993979.0,
      "step": 27470
    },
    {
      "entropy": 5.647634410858155,
      "epoch": 2.492289550072569,
      "grad_norm": 2.09375,
      "learning_rate": 0.00043786695033228604,
      "loss": 5.3159,
      "mean_token_accuracy": 0.17913970947265626,
      "num_tokens": 57004505.0,
      "step": 27475
    },
    {
      "entropy": 5.718615245819092,
      "epoch": 2.492743105950653,
      "grad_norm": 1.8125,
      "learning_rate": 0.00043784462420352747,
      "loss": 5.355,
      "mean_token_accuracy": 0.17944378554821014,
      "num_tokens": 57014121.0,
      "step": 27480
    },
    {
      "entropy": 5.6977180480957035,
      "epoch": 2.4931966618287373,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043782229470710067,
      "loss": 5.3187,
      "mean_token_accuracy": 0.17877397537231446,
      "num_tokens": 57023748.0,
      "step": 27485
    },
    {
      "entropy": 5.593630027770996,
      "epoch": 2.4936502177068216,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004377999618434673,
      "loss": 5.314,
      "mean_token_accuracy": 0.1763949826359749,
      "num_tokens": 57034782.0,
      "step": 27490
    },
    {
      "entropy": 5.598147344589234,
      "epoch": 2.4941037735849054,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004377776256130892,
      "loss": 5.2191,
      "mean_token_accuracy": 0.18398290425539016,
      "num_tokens": 57045136.0,
      "step": 27495
    },
    {
      "entropy": 5.68002028465271,
      "epoch": 2.4945573294629897,
      "grad_norm": 1.75,
      "learning_rate": 0.00043775528601642833,
      "loss": 5.4418,
      "mean_token_accuracy": 0.174806410074234,
      "num_tokens": 57055893.0,
      "step": 27500
    },
    {
      "entropy": 5.662081861495972,
      "epoch": 2.495010885341074,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043773294305394665,
      "loss": 5.3492,
      "mean_token_accuracy": 0.178668075799942,
      "num_tokens": 57067189.0,
      "step": 27505
    },
    {
      "entropy": 5.674722909927368,
      "epoch": 2.4954644412191582,
      "grad_norm": 1.5,
      "learning_rate": 0.00043771059672610636,
      "loss": 5.3425,
      "mean_token_accuracy": 0.18749552518129348,
      "num_tokens": 57077374.0,
      "step": 27510
    },
    {
      "entropy": 5.664820051193237,
      "epoch": 2.4959179970972425,
      "grad_norm": 1.78125,
      "learning_rate": 0.00043768824703336933,
      "loss": 5.3037,
      "mean_token_accuracy": 0.1768796995282173,
      "num_tokens": 57088168.0,
      "step": 27515
    },
    {
      "entropy": 5.637992572784424,
      "epoch": 2.4963715529753268,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004376658939761979,
      "loss": 5.2532,
      "mean_token_accuracy": 0.18334075510501863,
      "num_tokens": 57098743.0,
      "step": 27520
    },
    {
      "entropy": 5.603921031951904,
      "epoch": 2.4968251088534106,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004376435375550544,
      "loss": 5.4154,
      "mean_token_accuracy": 0.17553965151309966,
      "num_tokens": 57109854.0,
      "step": 27525
    },
    {
      "entropy": 5.607376956939698,
      "epoch": 2.497278664731495,
      "grad_norm": 3.546875,
      "learning_rate": 0.00043762117777040106,
      "loss": 5.2381,
      "mean_token_accuracy": 0.18721028715372084,
      "num_tokens": 57120296.0,
      "step": 27530
    },
    {
      "entropy": 5.636261129379273,
      "epoch": 2.497732220609579,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004375988146227003,
      "loss": 5.2778,
      "mean_token_accuracy": 0.18118964433670043,
      "num_tokens": 57131295.0,
      "step": 27535
    },
    {
      "entropy": 5.665420722961426,
      "epoch": 2.4981857764876634,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004375764481124147,
      "loss": 5.3476,
      "mean_token_accuracy": 0.1802725002169609,
      "num_tokens": 57141936.0,
      "step": 27540
    },
    {
      "entropy": 5.600212812423706,
      "epoch": 2.4986393323657476,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043755407824000664,
      "loss": 5.2817,
      "mean_token_accuracy": 0.18807021528482437,
      "num_tokens": 57152510.0,
      "step": 27545
    },
    {
      "entropy": 5.645239639282226,
      "epoch": 2.4990928882438315,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043753170500593884,
      "loss": 5.302,
      "mean_token_accuracy": 0.1821573868393898,
      "num_tokens": 57162546.0,
      "step": 27550
    },
    {
      "entropy": 5.595569086074829,
      "epoch": 2.4995464441219157,
      "grad_norm": 2.5,
      "learning_rate": 0.00043750932841067397,
      "loss": 5.243,
      "mean_token_accuracy": 0.18728336989879607,
      "num_tokens": 57174658.0,
      "step": 27555
    },
    {
      "entropy": 5.718395948410034,
      "epoch": 2.5,
      "grad_norm": 2.0625,
      "learning_rate": 0.00043748694845467483,
      "loss": 5.4434,
      "mean_token_accuracy": 0.1685873746871948,
      "num_tokens": 57185157.0,
      "step": 27560
    },
    {
      "entropy": 5.656469106674194,
      "epoch": 2.5004535558780843,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00043746456513840405,
      "loss": 5.215,
      "mean_token_accuracy": 0.18358564078807832,
      "num_tokens": 57195277.0,
      "step": 27565
    },
    {
      "entropy": 5.66600284576416,
      "epoch": 2.5009071117561685,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004374421784623248,
      "loss": 5.339,
      "mean_token_accuracy": 0.17541829198598863,
      "num_tokens": 57205374.0,
      "step": 27570
    },
    {
      "entropy": 5.642899799346924,
      "epoch": 2.501360667634253,
      "grad_norm": 1.875,
      "learning_rate": 0.00043741978842689987,
      "loss": 5.3121,
      "mean_token_accuracy": 0.17786130756139756,
      "num_tokens": 57215700.0,
      "step": 27575
    },
    {
      "entropy": 5.6669408798217775,
      "epoch": 2.5018142235123366,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004373973950325923,
      "loss": 5.3541,
      "mean_token_accuracy": 0.17943891286849975,
      "num_tokens": 57225765.0,
      "step": 27580
    },
    {
      "entropy": 5.687963438034058,
      "epoch": 2.502267779390421,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043737499827986527,
      "loss": 5.3256,
      "mean_token_accuracy": 0.18066371381282806,
      "num_tokens": 57235913.0,
      "step": 27585
    },
    {
      "entropy": 5.603995227813721,
      "epoch": 2.502721335268505,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043735259816918184,
      "loss": 5.2279,
      "mean_token_accuracy": 0.18346863985061646,
      "num_tokens": 57246204.0,
      "step": 27590
    },
    {
      "entropy": 5.601763343811035,
      "epoch": 2.5031748911465894,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004373301947010053,
      "loss": 5.327,
      "mean_token_accuracy": 0.1738245278596878,
      "num_tokens": 57256167.0,
      "step": 27595
    },
    {
      "entropy": 5.626915216445923,
      "epoch": 2.5036284470246732,
      "grad_norm": 1.75,
      "learning_rate": 0.00043730778787579895,
      "loss": 5.2904,
      "mean_token_accuracy": 0.18308962434530257,
      "num_tokens": 57266621.0,
      "step": 27600
    },
    {
      "entropy": 5.6757004737854,
      "epoch": 2.5040820029027575,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00043728537769402627,
      "loss": 5.3286,
      "mean_token_accuracy": 0.17583670765161513,
      "num_tokens": 57277282.0,
      "step": 27605
    },
    {
      "entropy": 5.66294493675232,
      "epoch": 2.5045355587808418,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004372629641561506,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1753845125436783,
      "num_tokens": 57287541.0,
      "step": 27610
    },
    {
      "entropy": 5.667787551879883,
      "epoch": 2.504989114658926,
      "grad_norm": 1.8125,
      "learning_rate": 0.00043724054726263547,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17684123814105987,
      "num_tokens": 57298158.0,
      "step": 27615
    },
    {
      "entropy": 5.654350280761719,
      "epoch": 2.5054426705370103,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004372181270139444,
      "loss": 5.322,
      "mean_token_accuracy": 0.1748050794005394,
      "num_tokens": 57308356.0,
      "step": 27620
    },
    {
      "entropy": 5.635295343399048,
      "epoch": 2.5058962264150946,
      "grad_norm": 2.625,
      "learning_rate": 0.00043719570341054127,
      "loss": 5.2428,
      "mean_token_accuracy": 0.18515816777944566,
      "num_tokens": 57317905.0,
      "step": 27625
    },
    {
      "entropy": 5.658818960189819,
      "epoch": 2.5063497822931784,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043717327645288956,
      "loss": 5.3704,
      "mean_token_accuracy": 0.18351224213838577,
      "num_tokens": 57328277.0,
      "step": 27630
    },
    {
      "entropy": 5.713400030136109,
      "epoch": 2.5068033381712627,
      "grad_norm": 1.625,
      "learning_rate": 0.00043715084614145316,
      "loss": 5.361,
      "mean_token_accuracy": 0.18127980679273606,
      "num_tokens": 57338975.0,
      "step": 27635
    },
    {
      "entropy": 5.6141444683074955,
      "epoch": 2.507256894049347,
      "grad_norm": 2.6875,
      "learning_rate": 0.000437128412476696,
      "loss": 5.2843,
      "mean_token_accuracy": 0.18473665565252304,
      "num_tokens": 57349439.0,
      "step": 27640
    },
    {
      "entropy": 5.62791953086853,
      "epoch": 2.507710449927431,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004371059754590819,
      "loss": 5.3634,
      "mean_token_accuracy": 0.18361365646123887,
      "num_tokens": 57359414.0,
      "step": 27645
    },
    {
      "entropy": 5.7775036811828615,
      "epoch": 2.508164005805515,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043708353508907494,
      "loss": 5.3951,
      "mean_token_accuracy": 0.17650662809610368,
      "num_tokens": 57369711.0,
      "step": 27650
    },
    {
      "entropy": 5.595389986038208,
      "epoch": 2.5086175616835993,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004370610913671392,
      "loss": 5.2304,
      "mean_token_accuracy": 0.19442549496889114,
      "num_tokens": 57380001.0,
      "step": 27655
    },
    {
      "entropy": 5.645458841323853,
      "epoch": 2.5090711175616836,
      "grad_norm": 4.3125,
      "learning_rate": 0.00043703864429373875,
      "loss": 5.411,
      "mean_token_accuracy": 0.17059244513511657,
      "num_tokens": 57390794.0,
      "step": 27660
    },
    {
      "entropy": 5.730699777603149,
      "epoch": 2.509524673439768,
      "grad_norm": 1.8125,
      "learning_rate": 0.00043701619386933793,
      "loss": 5.3601,
      "mean_token_accuracy": 0.17222311496734619,
      "num_tokens": 57401975.0,
      "step": 27665
    },
    {
      "entropy": 5.613577032089234,
      "epoch": 2.509978229317852,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043699374009440085,
      "loss": 5.2058,
      "mean_token_accuracy": 0.19298032820224761,
      "num_tokens": 57411249.0,
      "step": 27670
    },
    {
      "entropy": 5.585575723648072,
      "epoch": 2.5104317851959363,
      "grad_norm": 2.375,
      "learning_rate": 0.0004369712829693919,
      "loss": 5.2853,
      "mean_token_accuracy": 0.1888611525297165,
      "num_tokens": 57422679.0,
      "step": 27675
    },
    {
      "entropy": 5.628114366531372,
      "epoch": 2.51088534107402,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004369488224947757,
      "loss": 5.3109,
      "mean_token_accuracy": 0.18099258542060853,
      "num_tokens": 57433265.0,
      "step": 27680
    },
    {
      "entropy": 5.661364936828614,
      "epoch": 2.5113388969521044,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004369263586710164,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1813330426812172,
      "num_tokens": 57444095.0,
      "step": 27685
    },
    {
      "entropy": 5.6641796112060545,
      "epoch": 2.5117924528301887,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004369038914985789,
      "loss": 5.2704,
      "mean_token_accuracy": 0.18497756570577623,
      "num_tokens": 57454059.0,
      "step": 27690
    },
    {
      "entropy": 5.6321532249450685,
      "epoch": 2.512246008708273,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004368814209779276,
      "loss": 5.2611,
      "mean_token_accuracy": 0.1846875160932541,
      "num_tokens": 57463940.0,
      "step": 27695
    },
    {
      "entropy": 5.521158123016358,
      "epoch": 2.512699564586357,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00043685894710952724,
      "loss": 5.2963,
      "mean_token_accuracy": 0.17960627973079682,
      "num_tokens": 57474731.0,
      "step": 27700
    },
    {
      "entropy": 5.605391645431519,
      "epoch": 2.513153120464441,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004368364698938426,
      "loss": 5.2277,
      "mean_token_accuracy": 0.18872211277484893,
      "num_tokens": 57485096.0,
      "step": 27705
    },
    {
      "entropy": 5.636187791824341,
      "epoch": 2.5136066763425253,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004368139893313385,
      "loss": 5.3086,
      "mean_token_accuracy": 0.18104083091020584,
      "num_tokens": 57496371.0,
      "step": 27710
    },
    {
      "entropy": 5.688627910614014,
      "epoch": 2.5140602322206096,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043679150542248,
      "loss": 5.3141,
      "mean_token_accuracy": 0.1795603260397911,
      "num_tokens": 57506012.0,
      "step": 27715
    },
    {
      "entropy": 5.642393112182617,
      "epoch": 2.514513788098694,
      "grad_norm": 2.703125,
      "learning_rate": 0.00043676901816773174,
      "loss": 5.3515,
      "mean_token_accuracy": 0.17699277997016907,
      "num_tokens": 57516536.0,
      "step": 27720
    },
    {
      "entropy": 5.622779178619385,
      "epoch": 2.514967343976778,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004367465275675591,
      "loss": 5.2472,
      "mean_token_accuracy": 0.18471078723669052,
      "num_tokens": 57527450.0,
      "step": 27725
    },
    {
      "entropy": 5.704321050643921,
      "epoch": 2.5154208998548624,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00043672403362242704,
      "loss": 5.3738,
      "mean_token_accuracy": 0.18130140304565429,
      "num_tokens": 57538234.0,
      "step": 27730
    },
    {
      "entropy": 5.603015327453614,
      "epoch": 2.515874455732946,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00043670153633280066,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1857223019003868,
      "num_tokens": 57547938.0,
      "step": 27735
    },
    {
      "entropy": 5.677554035186768,
      "epoch": 2.5163280116110305,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004366790356991454,
      "loss": 5.3413,
      "mean_token_accuracy": 0.1811394765973091,
      "num_tokens": 57560553.0,
      "step": 27740
    },
    {
      "entropy": 5.635687160491943,
      "epoch": 2.5167815674891147,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004366565317219265,
      "loss": 5.2721,
      "mean_token_accuracy": 0.18375395089387894,
      "num_tokens": 57570670.0,
      "step": 27745
    },
    {
      "entropy": 5.726996898651123,
      "epoch": 2.5172351233671986,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043663402440160924,
      "loss": 5.4937,
      "mean_token_accuracy": 0.17109782695770265,
      "num_tokens": 57582377.0,
      "step": 27750
    },
    {
      "entropy": 5.634012508392334,
      "epoch": 2.517688679245283,
      "grad_norm": 1.25,
      "learning_rate": 0.0004366115137386592,
      "loss": 5.1957,
      "mean_token_accuracy": 0.1921502321958542,
      "num_tokens": 57592764.0,
      "step": 27755
    },
    {
      "entropy": 5.617668581008911,
      "epoch": 2.518142235123367,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004365889997335419,
      "loss": 5.2829,
      "mean_token_accuracy": 0.17624496668577194,
      "num_tokens": 57603768.0,
      "step": 27760
    },
    {
      "entropy": 5.629026126861572,
      "epoch": 2.5185957910014514,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00043656648238672285,
      "loss": 5.2871,
      "mean_token_accuracy": 0.18511344194412233,
      "num_tokens": 57614346.0,
      "step": 27765
    },
    {
      "entropy": 5.740295171737671,
      "epoch": 2.5190493468795356,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004365439616986678,
      "loss": 5.4404,
      "mean_token_accuracy": 0.1732534274458885,
      "num_tokens": 57625015.0,
      "step": 27770
    },
    {
      "entropy": 5.678911924362183,
      "epoch": 2.51950290275762,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043652143766984247,
      "loss": 5.378,
      "mean_token_accuracy": 0.1778662994503975,
      "num_tokens": 57635476.0,
      "step": 27775
    },
    {
      "entropy": 5.597596788406372,
      "epoch": 2.519956458635704,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043649891030071264,
      "loss": 5.2775,
      "mean_token_accuracy": 0.18269471675157548,
      "num_tokens": 57645512.0,
      "step": 27780
    },
    {
      "entropy": 5.615275955200195,
      "epoch": 2.520410014513788,
      "grad_norm": 2.28125,
      "learning_rate": 0.00043647637959174425,
      "loss": 5.2723,
      "mean_token_accuracy": 0.1772795245051384,
      "num_tokens": 57655766.0,
      "step": 27785
    },
    {
      "entropy": 5.669235563278198,
      "epoch": 2.5208635703918723,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043645384554340315,
      "loss": 5.3386,
      "mean_token_accuracy": 0.18109782189130783,
      "num_tokens": 57667325.0,
      "step": 27790
    },
    {
      "entropy": 5.626789283752442,
      "epoch": 2.5213171262699565,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043643130815615536,
      "loss": 5.2834,
      "mean_token_accuracy": 0.17715166062116622,
      "num_tokens": 57677884.0,
      "step": 27795
    },
    {
      "entropy": 5.711928176879883,
      "epoch": 2.521770682148041,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043640876743046694,
      "loss": 5.3352,
      "mean_token_accuracy": 0.18289560824632645,
      "num_tokens": 57687765.0,
      "step": 27800
    },
    {
      "entropy": 5.648309516906738,
      "epoch": 2.5222242380261246,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004363862233668041,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1770124539732933,
      "num_tokens": 57697777.0,
      "step": 27805
    },
    {
      "entropy": 5.6131508350372314,
      "epoch": 2.522677793904209,
      "grad_norm": 1.671875,
      "learning_rate": 0.000436363675965633,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17957866042852402,
      "num_tokens": 57708765.0,
      "step": 27810
    },
    {
      "entropy": 5.7117431640625,
      "epoch": 2.523131349782293,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004363411252274201,
      "loss": 5.3125,
      "mean_token_accuracy": 0.18514631539583207,
      "num_tokens": 57719658.0,
      "step": 27815
    },
    {
      "entropy": 5.698588895797729,
      "epoch": 2.5235849056603774,
      "grad_norm": 2.375,
      "learning_rate": 0.00043631857115263153,
      "loss": 5.3,
      "mean_token_accuracy": 0.1859341159462929,
      "num_tokens": 57730018.0,
      "step": 27820
    },
    {
      "entropy": 5.582456398010254,
      "epoch": 2.5240384615384617,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004362960137417338,
      "loss": 5.2386,
      "mean_token_accuracy": 0.1866927772760391,
      "num_tokens": 57741251.0,
      "step": 27825
    },
    {
      "entropy": 5.607759237289429,
      "epoch": 2.524492017416546,
      "grad_norm": 1.75,
      "learning_rate": 0.0004362734529951933,
      "loss": 5.3144,
      "mean_token_accuracy": 0.18160585612058638,
      "num_tokens": 57751020.0,
      "step": 27830
    },
    {
      "entropy": 5.6048877239227295,
      "epoch": 2.5249455732946298,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043625088891347685,
      "loss": 5.2799,
      "mean_token_accuracy": 0.1835087388753891,
      "num_tokens": 57760989.0,
      "step": 27835
    },
    {
      "entropy": 5.720717096328736,
      "epoch": 2.525399129172714,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043622832149705086,
      "loss": 5.354,
      "mean_token_accuracy": 0.18048861473798752,
      "num_tokens": 57770882.0,
      "step": 27840
    },
    {
      "entropy": 5.7173419952392575,
      "epoch": 2.5258526850507983,
      "grad_norm": 2.5625,
      "learning_rate": 0.00043620575074638206,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1785986080765724,
      "num_tokens": 57781812.0,
      "step": 27845
    },
    {
      "entropy": 5.653043937683106,
      "epoch": 2.5263062409288826,
      "grad_norm": 1.625,
      "learning_rate": 0.00043618317666193734,
      "loss": 5.29,
      "mean_token_accuracy": 0.1906338706612587,
      "num_tokens": 57792009.0,
      "step": 27850
    },
    {
      "entropy": 5.5958580493927,
      "epoch": 2.5267597968069664,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043616059924418346,
      "loss": 5.2615,
      "mean_token_accuracy": 0.19109372198581695,
      "num_tokens": 57802109.0,
      "step": 27855
    },
    {
      "entropy": 5.6406656265258786,
      "epoch": 2.5272133526850507,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004361380184935872,
      "loss": 5.3169,
      "mean_token_accuracy": 0.18603241443634033,
      "num_tokens": 57812374.0,
      "step": 27860
    },
    {
      "entropy": 5.640546560287476,
      "epoch": 2.527666908563135,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004361154344106158,
      "loss": 5.2202,
      "mean_token_accuracy": 0.18553358018398286,
      "num_tokens": 57822936.0,
      "step": 27865
    },
    {
      "entropy": 5.74988374710083,
      "epoch": 2.528120464441219,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043609284699573604,
      "loss": 5.4663,
      "mean_token_accuracy": 0.18017635941505433,
      "num_tokens": 57832693.0,
      "step": 27870
    },
    {
      "entropy": 5.70312533378601,
      "epoch": 2.5285740203193035,
      "grad_norm": 1.625,
      "learning_rate": 0.00043607025624941523,
      "loss": 5.332,
      "mean_token_accuracy": 0.18044215440750122,
      "num_tokens": 57843161.0,
      "step": 27875
    },
    {
      "entropy": 5.6345421314239506,
      "epoch": 2.5290275761973877,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004360476621721204,
      "loss": 5.2978,
      "mean_token_accuracy": 0.18876102268695832,
      "num_tokens": 57853027.0,
      "step": 27880
    },
    {
      "entropy": 5.6801494598388675,
      "epoch": 2.5294811320754715,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004360250647643189,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18756588250398637,
      "num_tokens": 57862821.0,
      "step": 27885
    },
    {
      "entropy": 5.638453245162964,
      "epoch": 2.529934687953556,
      "grad_norm": 1.78125,
      "learning_rate": 0.00043600246402647805,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18740108907222747,
      "num_tokens": 57874065.0,
      "step": 27890
    },
    {
      "entropy": 5.65429835319519,
      "epoch": 2.53038824383164,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004359798599590652,
      "loss": 5.3269,
      "mean_token_accuracy": 0.1809796780347824,
      "num_tokens": 57885190.0,
      "step": 27895
    },
    {
      "entropy": 5.619342708587647,
      "epoch": 2.5308417997097243,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043595725256254786,
      "loss": 5.2522,
      "mean_token_accuracy": 0.18260846734046937,
      "num_tokens": 57894234.0,
      "step": 27900
    },
    {
      "entropy": 5.656823825836182,
      "epoch": 2.531295355587808,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004359346418373935,
      "loss": 5.4101,
      "mean_token_accuracy": 0.1716233178973198,
      "num_tokens": 57905401.0,
      "step": 27905
    },
    {
      "entropy": 5.654559707641601,
      "epoch": 2.5317489114658924,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043591202778406975,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17868441045284272,
      "num_tokens": 57916097.0,
      "step": 27910
    },
    {
      "entropy": 5.569736289978027,
      "epoch": 2.5322024673439767,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004358894104030442,
      "loss": 5.181,
      "mean_token_accuracy": 0.18710199892520904,
      "num_tokens": 57926871.0,
      "step": 27915
    },
    {
      "entropy": 5.599502468109131,
      "epoch": 2.532656023222061,
      "grad_norm": 2.484375,
      "learning_rate": 0.00043586678969478464,
      "loss": 5.2564,
      "mean_token_accuracy": 0.19265828877687455,
      "num_tokens": 57937344.0,
      "step": 27920
    },
    {
      "entropy": 5.705892515182495,
      "epoch": 2.5331095791001452,
      "grad_norm": 1.578125,
      "learning_rate": 0.000435844165659759,
      "loss": 5.3731,
      "mean_token_accuracy": 0.17858865708112717,
      "num_tokens": 57947633.0,
      "step": 27925
    },
    {
      "entropy": 5.622838544845581,
      "epoch": 2.5335631349782295,
      "grad_norm": 2.5625,
      "learning_rate": 0.00043582153829843493,
      "loss": 5.2584,
      "mean_token_accuracy": 0.17880647629499435,
      "num_tokens": 57957948.0,
      "step": 27930
    },
    {
      "entropy": 5.680737495422363,
      "epoch": 2.5340166908563138,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004357989076112805,
      "loss": 5.3014,
      "mean_token_accuracy": 0.18251794576644897,
      "num_tokens": 57967719.0,
      "step": 27935
    },
    {
      "entropy": 5.7075995922088625,
      "epoch": 2.5344702467343976,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004357762735987637,
      "loss": 5.3709,
      "mean_token_accuracy": 0.17336725443601608,
      "num_tokens": 57978218.0,
      "step": 27940
    },
    {
      "entropy": 5.6132814407348635,
      "epoch": 2.534923802612482,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004357536362613525,
      "loss": 5.227,
      "mean_token_accuracy": 0.1887062594294548,
      "num_tokens": 57987860.0,
      "step": 27945
    },
    {
      "entropy": 5.65456748008728,
      "epoch": 2.535377358490566,
      "grad_norm": 1.78125,
      "learning_rate": 0.00043573099559951516,
      "loss": 5.3161,
      "mean_token_accuracy": 0.18763837665319444,
      "num_tokens": 57997872.0,
      "step": 27950
    },
    {
      "entropy": 5.696563959121704,
      "epoch": 2.5358309143686504,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004357083516137199,
      "loss": 5.3535,
      "mean_token_accuracy": 0.17981790453195573,
      "num_tokens": 58007771.0,
      "step": 27955
    },
    {
      "entropy": 5.709441995620727,
      "epoch": 2.536284470246734,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000435685704304435,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1735341027379036,
      "num_tokens": 58018972.0,
      "step": 27960
    },
    {
      "entropy": 5.6305821418762205,
      "epoch": 2.5367380261248185,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043566305367212884,
      "loss": 5.1935,
      "mean_token_accuracy": 0.18503215909004211,
      "num_tokens": 58028581.0,
      "step": 27965
    },
    {
      "entropy": 5.594999933242798,
      "epoch": 2.5371915820029027,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043564039971726964,
      "loss": 5.3313,
      "mean_token_accuracy": 0.1844057708978653,
      "num_tokens": 58038877.0,
      "step": 27970
    },
    {
      "entropy": 5.6750914573669435,
      "epoch": 2.537645137880987,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043561774244032614,
      "loss": 5.3064,
      "mean_token_accuracy": 0.1823980003595352,
      "num_tokens": 58048777.0,
      "step": 27975
    },
    {
      "entropy": 5.696079874038697,
      "epoch": 2.5380986937590713,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043559508184176675,
      "loss": 5.3784,
      "mean_token_accuracy": 0.1803280457854271,
      "num_tokens": 58060270.0,
      "step": 27980
    },
    {
      "entropy": 5.62990174293518,
      "epoch": 2.5385522496371555,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043557241792206014,
      "loss": 5.315,
      "mean_token_accuracy": 0.18501079976558685,
      "num_tokens": 58070838.0,
      "step": 27985
    },
    {
      "entropy": 5.684419250488281,
      "epoch": 2.5390058055152394,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043554975068167505,
      "loss": 5.3574,
      "mean_token_accuracy": 0.18224557042121886,
      "num_tokens": 58079933.0,
      "step": 27990
    },
    {
      "entropy": 5.656878471374512,
      "epoch": 2.5394593613933236,
      "grad_norm": 1.75,
      "learning_rate": 0.0004355270801210802,
      "loss": 5.3232,
      "mean_token_accuracy": 0.18315122425556182,
      "num_tokens": 58090673.0,
      "step": 27995
    },
    {
      "entropy": 5.602603387832642,
      "epoch": 2.539912917271408,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004355044062407444,
      "loss": 5.149,
      "mean_token_accuracy": 0.19133168309926987,
      "num_tokens": 58100172.0,
      "step": 28000
    },
    {
      "entropy": 5.531277084350586,
      "epoch": 2.540366473149492,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004354817290411366,
      "loss": 5.2974,
      "mean_token_accuracy": 0.18213654011487962,
      "num_tokens": 58108965.0,
      "step": 28005
    },
    {
      "entropy": 5.734698963165283,
      "epoch": 2.540820029027576,
      "grad_norm": 3.171875,
      "learning_rate": 0.00043545904852272573,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17504046112298965,
      "num_tokens": 58120094.0,
      "step": 28010
    },
    {
      "entropy": 5.755457162857056,
      "epoch": 2.5412735849056602,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00043543636468598083,
      "loss": 5.374,
      "mean_token_accuracy": 0.17513737976551055,
      "num_tokens": 58131525.0,
      "step": 28015
    },
    {
      "entropy": 5.607620906829834,
      "epoch": 2.5417271407837445,
      "grad_norm": 1.859375,
      "learning_rate": 0.00043541367753137107,
      "loss": 5.2793,
      "mean_token_accuracy": 0.18136280179023742,
      "num_tokens": 58141986.0,
      "step": 28020
    },
    {
      "entropy": 5.635508584976196,
      "epoch": 2.5421806966618288,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004353909870593656,
      "loss": 5.3414,
      "mean_token_accuracy": 0.18027782887220384,
      "num_tokens": 58154122.0,
      "step": 28025
    },
    {
      "entropy": 5.723494148254394,
      "epoch": 2.542634252539913,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004353682932704336,
      "loss": 5.2447,
      "mean_token_accuracy": 0.19322414994239806,
      "num_tokens": 58164148.0,
      "step": 28030
    },
    {
      "entropy": 5.576816987991333,
      "epoch": 2.5430878084179973,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004353455961650444,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18285716325044632,
      "num_tokens": 58173836.0,
      "step": 28035
    },
    {
      "entropy": 5.578534460067749,
      "epoch": 2.543541364296081,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004353228957436675,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18462584614753724,
      "num_tokens": 58184598.0,
      "step": 28040
    },
    {
      "entropy": 5.5955140590667725,
      "epoch": 2.5439949201741654,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00043530019200677215,
      "loss": 5.2635,
      "mean_token_accuracy": 0.18238011598587037,
      "num_tokens": 58194820.0,
      "step": 28045
    },
    {
      "entropy": 5.670554304122925,
      "epoch": 2.5444484760522497,
      "grad_norm": 2.03125,
      "learning_rate": 0.00043527748495482806,
      "loss": 5.3797,
      "mean_token_accuracy": 0.17689321041107178,
      "num_tokens": 58206589.0,
      "step": 28050
    },
    {
      "entropy": 5.706000328063965,
      "epoch": 2.544902031930334,
      "grad_norm": 2.03125,
      "learning_rate": 0.00043525477458830473,
      "loss": 5.3077,
      "mean_token_accuracy": 0.182235948741436,
      "num_tokens": 58216750.0,
      "step": 28055
    },
    {
      "entropy": 5.610604953765869,
      "epoch": 2.5453555878084178,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004352320609076718,
      "loss": 5.2978,
      "mean_token_accuracy": 0.18005290776491165,
      "num_tokens": 58228496.0,
      "step": 28060
    },
    {
      "entropy": 5.599316024780274,
      "epoch": 2.545809143686502,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043520934391339894,
      "loss": 5.2396,
      "mean_token_accuracy": 0.1842537209391594,
      "num_tokens": 58237844.0,
      "step": 28065
    },
    {
      "entropy": 5.633706426620483,
      "epoch": 2.5462626995645863,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004351866236059561,
      "loss": 5.2401,
      "mean_token_accuracy": 0.18382677882909776,
      "num_tokens": 58248522.0,
      "step": 28070
    },
    {
      "entropy": 5.627744150161743,
      "epoch": 2.5467162554426706,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00043516389998581305,
      "loss": 5.2381,
      "mean_token_accuracy": 0.18606381863355637,
      "num_tokens": 58258492.0,
      "step": 28075
    },
    {
      "entropy": 5.526512479782104,
      "epoch": 2.547169811320755,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004351411730534398,
      "loss": 5.2783,
      "mean_token_accuracy": 0.17916622012853622,
      "num_tokens": 58268683.0,
      "step": 28080
    },
    {
      "entropy": 5.664614534378051,
      "epoch": 2.547623367198839,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004351184428093062,
      "loss": 5.3671,
      "mean_token_accuracy": 0.177337546646595,
      "num_tokens": 58279479.0,
      "step": 28085
    },
    {
      "entropy": 5.745540904998779,
      "epoch": 2.5480769230769234,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004350957092538825,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17782476991415025,
      "num_tokens": 58290751.0,
      "step": 28090
    },
    {
      "entropy": 5.689977931976318,
      "epoch": 2.548530478955007,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004350729723876386,
      "loss": 5.3708,
      "mean_token_accuracy": 0.17747713774442672,
      "num_tokens": 58302163.0,
      "step": 28095
    },
    {
      "entropy": 5.608311939239502,
      "epoch": 2.5489840348330914,
      "grad_norm": 2.1875,
      "learning_rate": 0.00043505023221104497,
      "loss": 5.314,
      "mean_token_accuracy": 0.1785145655274391,
      "num_tokens": 58311488.0,
      "step": 28100
    },
    {
      "entropy": 5.653325843811035,
      "epoch": 2.5494375907111757,
      "grad_norm": 2.953125,
      "learning_rate": 0.00043502748872457166,
      "loss": 5.2803,
      "mean_token_accuracy": 0.18149032145738603,
      "num_tokens": 58321599.0,
      "step": 28105
    },
    {
      "entropy": 5.629932117462158,
      "epoch": 2.5498911465892595,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004350047419286892,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18210791200399398,
      "num_tokens": 58330696.0,
      "step": 28110
    },
    {
      "entropy": 5.590297746658325,
      "epoch": 2.550344702467344,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043498199182386775,
      "loss": 5.3234,
      "mean_token_accuracy": 0.18329794853925704,
      "num_tokens": 58341793.0,
      "step": 28115
    },
    {
      "entropy": 5.6936944961547855,
      "epoch": 2.550798258345428,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00043495923841057807,
      "loss": 5.3889,
      "mean_token_accuracy": 0.17490618973970412,
      "num_tokens": 58352260.0,
      "step": 28120
    },
    {
      "entropy": 5.756526899337769,
      "epoch": 2.5512518142235123,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043493648168929053,
      "loss": 5.4402,
      "mean_token_accuracy": 0.16971493661403655,
      "num_tokens": 58363052.0,
      "step": 28125
    },
    {
      "entropy": 5.725798320770264,
      "epoch": 2.5517053701015966,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004349137216604758,
      "loss": 5.3326,
      "mean_token_accuracy": 0.1773318812251091,
      "num_tokens": 58374039.0,
      "step": 28130
    },
    {
      "entropy": 5.576448535919189,
      "epoch": 2.552158925979681,
      "grad_norm": 1.75,
      "learning_rate": 0.0004348909583246045,
      "loss": 5.1492,
      "mean_token_accuracy": 0.19279193878173828,
      "num_tokens": 58383441.0,
      "step": 28135
    },
    {
      "entropy": 5.56383318901062,
      "epoch": 2.552612481857765,
      "grad_norm": 1.375,
      "learning_rate": 0.0004348681916821475,
      "loss": 5.2669,
      "mean_token_accuracy": 0.18274327218532563,
      "num_tokens": 58394164.0,
      "step": 28140
    },
    {
      "entropy": 5.632232570648194,
      "epoch": 2.553066037735849,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004348454217335755,
      "loss": 5.2884,
      "mean_token_accuracy": 0.18097569048404694,
      "num_tokens": 58404610.0,
      "step": 28145
    },
    {
      "entropy": 5.61922607421875,
      "epoch": 2.553519593613933,
      "grad_norm": 2.8125,
      "learning_rate": 0.00043482264847935953,
      "loss": 5.2065,
      "mean_token_accuracy": 0.18763020187616347,
      "num_tokens": 58414512.0,
      "step": 28150
    },
    {
      "entropy": 5.659669876098633,
      "epoch": 2.5539731494920175,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043479987191997045,
      "loss": 5.3939,
      "mean_token_accuracy": 0.17854946106672287,
      "num_tokens": 58424669.0,
      "step": 28155
    },
    {
      "entropy": 5.582615089416504,
      "epoch": 2.5544267053701017,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004347770920558792,
      "loss": 5.2196,
      "mean_token_accuracy": 0.18647691756486892,
      "num_tokens": 58435348.0,
      "step": 28160
    },
    {
      "entropy": 5.64633321762085,
      "epoch": 2.5548802612481856,
      "grad_norm": 2.0,
      "learning_rate": 0.00043475430888755706,
      "loss": 5.2877,
      "mean_token_accuracy": 0.17828176468610762,
      "num_tokens": 58445534.0,
      "step": 28165
    },
    {
      "entropy": 5.659234714508057,
      "epoch": 2.55533381712627,
      "grad_norm": 2.125,
      "learning_rate": 0.0004347315224154751,
      "loss": 5.33,
      "mean_token_accuracy": 0.18123072981834412,
      "num_tokens": 58456718.0,
      "step": 28170
    },
    {
      "entropy": 5.7816873550415036,
      "epoch": 2.555787373004354,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004347087326401045,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17642312347888947,
      "num_tokens": 58467176.0,
      "step": 28175
    },
    {
      "entropy": 5.603309154510498,
      "epoch": 2.5562409288824384,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00043468593956191663,
      "loss": 5.2326,
      "mean_token_accuracy": 0.18815156072378159,
      "num_tokens": 58477642.0,
      "step": 28180
    },
    {
      "entropy": 5.579352998733521,
      "epoch": 2.5566944847605226,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004346631431813829,
      "loss": 5.2787,
      "mean_token_accuracy": 0.17970601618289947,
      "num_tokens": 58489055.0,
      "step": 28185
    },
    {
      "entropy": 5.68219895362854,
      "epoch": 2.557148040638607,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004346403434989746,
      "loss": 5.2955,
      "mean_token_accuracy": 0.18244461864233016,
      "num_tokens": 58499062.0,
      "step": 28190
    },
    {
      "entropy": 5.647238254547119,
      "epoch": 2.5576015965166907,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004346175405151633,
      "loss": 5.2979,
      "mean_token_accuracy": 0.1779731974005699,
      "num_tokens": 58509154.0,
      "step": 28195
    },
    {
      "entropy": 5.623441457748413,
      "epoch": 2.558055152394775,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004345947342304207,
      "loss": 5.2204,
      "mean_token_accuracy": 0.18578416258096694,
      "num_tokens": 58520056.0,
      "step": 28200
    },
    {
      "entropy": 5.640213394165039,
      "epoch": 2.5585087082728593,
      "grad_norm": 1.75,
      "learning_rate": 0.00043457192464521827,
      "loss": 5.2841,
      "mean_token_accuracy": 0.1851004034280777,
      "num_tokens": 58529934.0,
      "step": 28205
    },
    {
      "entropy": 5.596674966812134,
      "epoch": 2.5589622641509435,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004345491117600278,
      "loss": 5.2128,
      "mean_token_accuracy": 0.1886480301618576,
      "num_tokens": 58540013.0,
      "step": 28210
    },
    {
      "entropy": 5.722359037399292,
      "epoch": 2.5594158200290273,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043452629557532097,
      "loss": 5.3123,
      "mean_token_accuracy": 0.1806127682328224,
      "num_tokens": 58550282.0,
      "step": 28215
    },
    {
      "entropy": 5.625477361679077,
      "epoch": 2.5598693759071116,
      "grad_norm": 2.25,
      "learning_rate": 0.00043450347609156974,
      "loss": 5.3384,
      "mean_token_accuracy": 0.1722565174102783,
      "num_tokens": 58560490.0,
      "step": 28220
    },
    {
      "entropy": 5.622791194915772,
      "epoch": 2.560322931785196,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043448065330924595,
      "loss": 5.335,
      "mean_token_accuracy": 0.1771353855729103,
      "num_tokens": 58570371.0,
      "step": 28225
    },
    {
      "entropy": 5.728695821762085,
      "epoch": 2.56077648766328,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004344578272288216,
      "loss": 5.2723,
      "mean_token_accuracy": 0.18708024322986602,
      "num_tokens": 58580701.0,
      "step": 28230
    },
    {
      "entropy": 5.631698369979858,
      "epoch": 2.5612300435413644,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043443499785076885,
      "loss": 5.3344,
      "mean_token_accuracy": 0.17854304164648055,
      "num_tokens": 58590903.0,
      "step": 28235
    },
    {
      "entropy": 5.6593403816223145,
      "epoch": 2.5616835994194487,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004344121651755596,
      "loss": 5.3131,
      "mean_token_accuracy": 0.17269703596830369,
      "num_tokens": 58600924.0,
      "step": 28240
    },
    {
      "entropy": 5.681178331375122,
      "epoch": 2.5621371552975325,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043438932920366615,
      "loss": 5.3737,
      "mean_token_accuracy": 0.18282488286495208,
      "num_tokens": 58611369.0,
      "step": 28245
    },
    {
      "entropy": 5.745564079284668,
      "epoch": 2.5625907111756168,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004343664899355608,
      "loss": 5.3707,
      "mean_token_accuracy": 0.182684825360775,
      "num_tokens": 58622725.0,
      "step": 28250
    },
    {
      "entropy": 5.663812112808228,
      "epoch": 2.563044267053701,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00043434364737171587,
      "loss": 5.3212,
      "mean_token_accuracy": 0.18165679275989532,
      "num_tokens": 58634157.0,
      "step": 28255
    },
    {
      "entropy": 5.597874021530151,
      "epoch": 2.5634978229317853,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043432080151260366,
      "loss": 5.1963,
      "mean_token_accuracy": 0.1828600913286209,
      "num_tokens": 58643384.0,
      "step": 28260
    },
    {
      "entropy": 5.631106853485107,
      "epoch": 2.563951378809869,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043429795235869667,
      "loss": 5.2743,
      "mean_token_accuracy": 0.1791883498430252,
      "num_tokens": 58653748.0,
      "step": 28265
    },
    {
      "entropy": 5.67836709022522,
      "epoch": 2.5644049346879534,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043427509991046746,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17933699786663054,
      "num_tokens": 58664619.0,
      "step": 28270
    },
    {
      "entropy": 5.623795509338379,
      "epoch": 2.5648584905660377,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043425224416838857,
      "loss": 5.2118,
      "mean_token_accuracy": 0.19191750586032869,
      "num_tokens": 58674506.0,
      "step": 28275
    },
    {
      "entropy": 5.674460506439209,
      "epoch": 2.565312046444122,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004342293851329327,
      "loss": 5.3831,
      "mean_token_accuracy": 0.17489042580127717,
      "num_tokens": 58686252.0,
      "step": 28280
    },
    {
      "entropy": 5.673289394378662,
      "epoch": 2.565765602322206,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043420652280457255,
      "loss": 5.3002,
      "mean_token_accuracy": 0.17801942974328994,
      "num_tokens": 58696477.0,
      "step": 28285
    },
    {
      "entropy": 5.512887191772461,
      "epoch": 2.5662191582002905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043418365718378096,
      "loss": 5.1571,
      "mean_token_accuracy": 0.19080700129270553,
      "num_tokens": 58706664.0,
      "step": 28290
    },
    {
      "entropy": 5.565709829330444,
      "epoch": 2.5666727140783747,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004341607882710307,
      "loss": 5.2459,
      "mean_token_accuracy": 0.1851401910185814,
      "num_tokens": 58716950.0,
      "step": 28295
    },
    {
      "entropy": 5.630371809005737,
      "epoch": 2.5671262699564585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004341379160667949,
      "loss": 5.2864,
      "mean_token_accuracy": 0.18106427192687988,
      "num_tokens": 58727723.0,
      "step": 28300
    },
    {
      "entropy": 5.592650651931763,
      "epoch": 2.567579825834543,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004341150405715464,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17647855579853058,
      "num_tokens": 58737956.0,
      "step": 28305
    },
    {
      "entropy": 5.642541551589966,
      "epoch": 2.568033381712627,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043409216178575826,
      "loss": 5.2305,
      "mean_token_accuracy": 0.1960057482123375,
      "num_tokens": 58748476.0,
      "step": 28310
    },
    {
      "entropy": 5.674935960769654,
      "epoch": 2.5684869375907113,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004340692797099037,
      "loss": 5.3364,
      "mean_token_accuracy": 0.18430718928575515,
      "num_tokens": 58759604.0,
      "step": 28315
    },
    {
      "entropy": 5.659968137741089,
      "epoch": 2.568940493468795,
      "grad_norm": 1.5,
      "learning_rate": 0.0004340463943444559,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18304712623357772,
      "num_tokens": 58769781.0,
      "step": 28320
    },
    {
      "entropy": 5.698338890075684,
      "epoch": 2.5693940493468794,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004340235056898882,
      "loss": 5.3333,
      "mean_token_accuracy": 0.182713682949543,
      "num_tokens": 58779384.0,
      "step": 28325
    },
    {
      "entropy": 5.719322156906128,
      "epoch": 2.5698476052249637,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00043400061374667374,
      "loss": 5.4047,
      "mean_token_accuracy": 0.1779683366417885,
      "num_tokens": 58791953.0,
      "step": 28330
    },
    {
      "entropy": 5.600538349151611,
      "epoch": 2.570301161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004339777185152862,
      "loss": 5.2921,
      "mean_token_accuracy": 0.18008000701665877,
      "num_tokens": 58802415.0,
      "step": 28335
    },
    {
      "entropy": 5.720829105377197,
      "epoch": 2.5707547169811322,
      "grad_norm": 1.625,
      "learning_rate": 0.0004339548199961989,
      "loss": 5.3083,
      "mean_token_accuracy": 0.1837185427546501,
      "num_tokens": 58812582.0,
      "step": 28340
    },
    {
      "entropy": 5.65279598236084,
      "epoch": 2.5712082728592165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043393191818988543,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17840672433376312,
      "num_tokens": 58822987.0,
      "step": 28345
    },
    {
      "entropy": 5.620814037322998,
      "epoch": 2.5716618287373003,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004339090130968193,
      "loss": 5.2425,
      "mean_token_accuracy": 0.18902881294488907,
      "num_tokens": 58832651.0,
      "step": 28350
    },
    {
      "entropy": 5.643963813781738,
      "epoch": 2.5721153846153846,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004338861047174743,
      "loss": 5.3192,
      "mean_token_accuracy": 0.17786250114440919,
      "num_tokens": 58842493.0,
      "step": 28355
    },
    {
      "entropy": 5.656501388549804,
      "epoch": 2.572568940493469,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004338631930523242,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1866173192858696,
      "num_tokens": 58852428.0,
      "step": 28360
    },
    {
      "entropy": 5.590909337997436,
      "epoch": 2.573022496371553,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004338402781018429,
      "loss": 5.2262,
      "mean_token_accuracy": 0.19385078102350234,
      "num_tokens": 58863234.0,
      "step": 28365
    },
    {
      "entropy": 5.559344911575318,
      "epoch": 2.573476052249637,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000433817359866504,
      "loss": 5.2503,
      "mean_token_accuracy": 0.1888156235218048,
      "num_tokens": 58872632.0,
      "step": 28370
    },
    {
      "entropy": 5.679463624954224,
      "epoch": 2.573929608127721,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004337944383467817,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1681532233953476,
      "num_tokens": 58882774.0,
      "step": 28375
    },
    {
      "entropy": 5.654532957077026,
      "epoch": 2.5743831640058055,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043377151354314996,
      "loss": 5.2299,
      "mean_token_accuracy": 0.19357341676950454,
      "num_tokens": 58892380.0,
      "step": 28380
    },
    {
      "entropy": 5.738065338134765,
      "epoch": 2.5748367198838897,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004337485854560829,
      "loss": 5.2973,
      "mean_token_accuracy": 0.18113393783569337,
      "num_tokens": 58902659.0,
      "step": 28385
    },
    {
      "entropy": 5.602086210250855,
      "epoch": 2.575290275761974,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004337256540860545,
      "loss": 5.2622,
      "mean_token_accuracy": 0.18372650742530822,
      "num_tokens": 58913979.0,
      "step": 28390
    },
    {
      "entropy": 5.668359422683716,
      "epoch": 2.5757438316400583,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004337027194335393,
      "loss": 5.364,
      "mean_token_accuracy": 0.18262268006801605,
      "num_tokens": 58924481.0,
      "step": 28395
    },
    {
      "entropy": 5.668369913101197,
      "epoch": 2.576197387518142,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00043367978149901125,
      "loss": 5.3222,
      "mean_token_accuracy": 0.17759879976511,
      "num_tokens": 58936287.0,
      "step": 28400
    },
    {
      "entropy": 5.6688563346862795,
      "epoch": 2.5766509433962264,
      "grad_norm": 1.84375,
      "learning_rate": 0.00043365684028294493,
      "loss": 5.3041,
      "mean_token_accuracy": 0.18284576684236525,
      "num_tokens": 58946070.0,
      "step": 28405
    },
    {
      "entropy": 5.666455078125,
      "epoch": 2.5771044992743106,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043363389578581483,
      "loss": 5.3478,
      "mean_token_accuracy": 0.17485314160585402,
      "num_tokens": 58956083.0,
      "step": 28410
    },
    {
      "entropy": 5.686349582672119,
      "epoch": 2.577558055152395,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004336109480080953,
      "loss": 5.3113,
      "mean_token_accuracy": 0.17424700707197188,
      "num_tokens": 58967557.0,
      "step": 28415
    },
    {
      "entropy": 5.598721361160278,
      "epoch": 2.5780116110304787,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004335879969502608,
      "loss": 5.2534,
      "mean_token_accuracy": 0.1835335060954094,
      "num_tokens": 58977509.0,
      "step": 28420
    },
    {
      "entropy": 5.523345470428467,
      "epoch": 2.578465166908563,
      "grad_norm": 2.078125,
      "learning_rate": 0.00043356504261278625,
      "loss": 5.1621,
      "mean_token_accuracy": 0.1962965652346611,
      "num_tokens": 58987498.0,
      "step": 28425
    },
    {
      "entropy": 5.711716651916504,
      "epoch": 2.5789187227866472,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004335420849961462,
      "loss": 5.4112,
      "mean_token_accuracy": 0.16945041418075563,
      "num_tokens": 58998503.0,
      "step": 28430
    },
    {
      "entropy": 5.62449083328247,
      "epoch": 2.5793722786647315,
      "grad_norm": 1.625,
      "learning_rate": 0.0004335191241008154,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17961693108081817,
      "num_tokens": 59007384.0,
      "step": 28435
    },
    {
      "entropy": 5.700660943984985,
      "epoch": 2.5798258345428158,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004334961599272687,
      "loss": 5.4128,
      "mean_token_accuracy": 0.17769681364297868,
      "num_tokens": 59017739.0,
      "step": 28440
    },
    {
      "entropy": 5.705712175369262,
      "epoch": 2.5802793904209,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004334731924759811,
      "loss": 5.337,
      "mean_token_accuracy": 0.1842995911836624,
      "num_tokens": 59028439.0,
      "step": 28445
    },
    {
      "entropy": 5.670135498046875,
      "epoch": 2.5807329462989843,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004334502217474274,
      "loss": 5.2603,
      "mean_token_accuracy": 0.1812853381037712,
      "num_tokens": 59039250.0,
      "step": 28450
    },
    {
      "entropy": 5.717446136474609,
      "epoch": 2.581186502177068,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043342724774208277,
      "loss": 5.3827,
      "mean_token_accuracy": 0.18083306699991225,
      "num_tokens": 59049375.0,
      "step": 28455
    },
    {
      "entropy": 5.6713964462280275,
      "epoch": 2.5816400580551524,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043340427046042237,
      "loss": 5.3832,
      "mean_token_accuracy": 0.1820751503109932,
      "num_tokens": 59059873.0,
      "step": 28460
    },
    {
      "entropy": 5.621150493621826,
      "epoch": 2.5820936139332367,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004333812899029212,
      "loss": 5.259,
      "mean_token_accuracy": 0.18614727705717088,
      "num_tokens": 59069288.0,
      "step": 28465
    },
    {
      "entropy": 5.642261266708374,
      "epoch": 2.5825471698113205,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043335830607005456,
      "loss": 5.2786,
      "mean_token_accuracy": 0.18170377016067504,
      "num_tokens": 59079339.0,
      "step": 28470
    },
    {
      "entropy": 5.695300626754761,
      "epoch": 2.5830007256894048,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004333353189622979,
      "loss": 5.3083,
      "mean_token_accuracy": 0.1772730752825737,
      "num_tokens": 59089883.0,
      "step": 28475
    },
    {
      "entropy": 5.675390577316284,
      "epoch": 2.583454281567489,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004333123285801264,
      "loss": 5.402,
      "mean_token_accuracy": 0.18456652462482454,
      "num_tokens": 59100532.0,
      "step": 28480
    },
    {
      "entropy": 5.638532638549805,
      "epoch": 2.5839078374455733,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004332893349240156,
      "loss": 5.2827,
      "mean_token_accuracy": 0.1829397663474083,
      "num_tokens": 59111295.0,
      "step": 28485
    },
    {
      "entropy": 5.668371200561523,
      "epoch": 2.5843613933236576,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004332663379944411,
      "loss": 5.2982,
      "mean_token_accuracy": 0.18155535012483598,
      "num_tokens": 59120808.0,
      "step": 28490
    },
    {
      "entropy": 5.720349645614624,
      "epoch": 2.584814949201742,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043324333779187844,
      "loss": 5.3493,
      "mean_token_accuracy": 0.19004903584718705,
      "num_tokens": 59131438.0,
      "step": 28495
    },
    {
      "entropy": 5.575861835479737,
      "epoch": 2.585268505079826,
      "grad_norm": 2.0,
      "learning_rate": 0.00043322033431680314,
      "loss": 5.3012,
      "mean_token_accuracy": 0.18981255292892457,
      "num_tokens": 59141701.0,
      "step": 28500
    },
    {
      "entropy": 5.589477586746216,
      "epoch": 2.58572206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.000433197327569691,
      "loss": 5.2779,
      "mean_token_accuracy": 0.1875492364168167,
      "num_tokens": 59152184.0,
      "step": 28505
    },
    {
      "entropy": 5.651774740219116,
      "epoch": 2.586175616835994,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043317431755101785,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1934633418917656,
      "num_tokens": 59162177.0,
      "step": 28510
    },
    {
      "entropy": 5.695134162902832,
      "epoch": 2.5866291727140784,
      "grad_norm": 1.625,
      "learning_rate": 0.0004331513042612594,
      "loss": 5.383,
      "mean_token_accuracy": 0.1852618247270584,
      "num_tokens": 59171964.0,
      "step": 28515
    },
    {
      "entropy": 5.628999137878418,
      "epoch": 2.5870827285921627,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004331282877008918,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1815845012664795,
      "num_tokens": 59182224.0,
      "step": 28520
    },
    {
      "entropy": 5.660320043563843,
      "epoch": 2.5875362844702465,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004331052678703909,
      "loss": 5.2884,
      "mean_token_accuracy": 0.18208315670490266,
      "num_tokens": 59191912.0,
      "step": 28525
    },
    {
      "entropy": 5.668180322647094,
      "epoch": 2.587989840348331,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004330822447702327,
      "loss": 5.251,
      "mean_token_accuracy": 0.18311957865953446,
      "num_tokens": 59200984.0,
      "step": 28530
    },
    {
      "entropy": 5.6906664848327635,
      "epoch": 2.588443396226415,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043305921840089347,
      "loss": 5.3378,
      "mean_token_accuracy": 0.18066257536411284,
      "num_tokens": 59210672.0,
      "step": 28535
    },
    {
      "entropy": 5.644884967803955,
      "epoch": 2.5888969521044993,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043303618876284935,
      "loss": 5.3129,
      "mean_token_accuracy": 0.18731480985879898,
      "num_tokens": 59220230.0,
      "step": 28540
    },
    {
      "entropy": 5.575934648513794,
      "epoch": 2.5893505079825836,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043301315585657645,
      "loss": 5.2295,
      "mean_token_accuracy": 0.18409752994775772,
      "num_tokens": 59230615.0,
      "step": 28545
    },
    {
      "entropy": 5.7064817428588865,
      "epoch": 2.589804063860668,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004329901196825513,
      "loss": 5.3521,
      "mean_token_accuracy": 0.18307065665721894,
      "num_tokens": 59241717.0,
      "step": 28550
    },
    {
      "entropy": 5.732670688629151,
      "epoch": 2.5902576197387517,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043296708024125025,
      "loss": 5.3427,
      "mean_token_accuracy": 0.18135012537240983,
      "num_tokens": 59251791.0,
      "step": 28555
    },
    {
      "entropy": 5.5845465660095215,
      "epoch": 2.590711175616836,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043294403753314965,
      "loss": 5.1869,
      "mean_token_accuracy": 0.19015798568725586,
      "num_tokens": 59262434.0,
      "step": 28560
    },
    {
      "entropy": 5.577397584915161,
      "epoch": 2.59116473149492,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004329209915587261,
      "loss": 5.3415,
      "mean_token_accuracy": 0.18083159476518632,
      "num_tokens": 59272138.0,
      "step": 28565
    },
    {
      "entropy": 5.670657587051392,
      "epoch": 2.5916182873730045,
      "grad_norm": 2.0,
      "learning_rate": 0.00043289794231845624,
      "loss": 5.3535,
      "mean_token_accuracy": 0.17904478311538696,
      "num_tokens": 59283470.0,
      "step": 28570
    },
    {
      "entropy": 5.619774723052979,
      "epoch": 2.5920718432510883,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004328748898128168,
      "loss": 5.2903,
      "mean_token_accuracy": 0.18602294325828553,
      "num_tokens": 59293670.0,
      "step": 28575
    },
    {
      "entropy": 5.6504110336303714,
      "epoch": 2.5925253991291726,
      "grad_norm": 2.125,
      "learning_rate": 0.0004328518340422842,
      "loss": 5.2956,
      "mean_token_accuracy": 0.1850317597389221,
      "num_tokens": 59303728.0,
      "step": 28580
    },
    {
      "entropy": 5.592752742767334,
      "epoch": 2.592978955007257,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043282877500733555,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18609225898981094,
      "num_tokens": 59314271.0,
      "step": 28585
    },
    {
      "entropy": 5.626435041427612,
      "epoch": 2.593432510885341,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004328057127084476,
      "loss": 5.304,
      "mean_token_accuracy": 0.18300392776727675,
      "num_tokens": 59324581.0,
      "step": 28590
    },
    {
      "entropy": 5.625092697143555,
      "epoch": 2.5938860667634254,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043278264714609733,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18879603445529938,
      "num_tokens": 59333912.0,
      "step": 28595
    },
    {
      "entropy": 5.610384225845337,
      "epoch": 2.5943396226415096,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043275957832076173,
      "loss": 5.2362,
      "mean_token_accuracy": 0.18306912183761598,
      "num_tokens": 59344245.0,
      "step": 28600
    },
    {
      "entropy": 5.615690517425537,
      "epoch": 2.5947931785195935,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00043273650623291775,
      "loss": 5.3627,
      "mean_token_accuracy": 0.18531100153923036,
      "num_tokens": 59353825.0,
      "step": 28605
    },
    {
      "entropy": 5.691075944900513,
      "epoch": 2.5952467343976777,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004327134308830427,
      "loss": 5.3254,
      "mean_token_accuracy": 0.17847936302423478,
      "num_tokens": 59364901.0,
      "step": 28610
    },
    {
      "entropy": 5.5847673416137695,
      "epoch": 2.595700290275762,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004326903522716137,
      "loss": 5.2664,
      "mean_token_accuracy": 0.18192201405763625,
      "num_tokens": 59374772.0,
      "step": 28615
    },
    {
      "entropy": 5.511993312835694,
      "epoch": 2.5961538461538463,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000432667270399108,
      "loss": 5.1628,
      "mean_token_accuracy": 0.20111260265111924,
      "num_tokens": 59383640.0,
      "step": 28620
    },
    {
      "entropy": 5.611902809143066,
      "epoch": 2.59660740203193,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043264418526600305,
      "loss": 5.2597,
      "mean_token_accuracy": 0.18298691064119338,
      "num_tokens": 59392928.0,
      "step": 28625
    },
    {
      "entropy": 5.743429279327392,
      "epoch": 2.5970609579100143,
      "grad_norm": 1.375,
      "learning_rate": 0.0004326210968727761,
      "loss": 5.3469,
      "mean_token_accuracy": 0.17737664878368378,
      "num_tokens": 59402969.0,
      "step": 28630
    },
    {
      "entropy": 5.675991201400757,
      "epoch": 2.5975145137880986,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004325980052199047,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17552854269742965,
      "num_tokens": 59413439.0,
      "step": 28635
    },
    {
      "entropy": 5.665130615234375,
      "epoch": 2.597968069666183,
      "grad_norm": 1.375,
      "learning_rate": 0.00043257491030786644,
      "loss": 5.3431,
      "mean_token_accuracy": 0.1799698993563652,
      "num_tokens": 59423403.0,
      "step": 28640
    },
    {
      "entropy": 5.728023338317871,
      "epoch": 2.598421625544267,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043255181213713883,
      "loss": 5.39,
      "mean_token_accuracy": 0.1765243664383888,
      "num_tokens": 59434519.0,
      "step": 28645
    },
    {
      "entropy": 5.6759583950042725,
      "epoch": 2.5988751814223514,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004325287107081996,
      "loss": 5.2972,
      "mean_token_accuracy": 0.1777392417192459,
      "num_tokens": 59445056.0,
      "step": 28650
    },
    {
      "entropy": 5.598626136779785,
      "epoch": 2.5993287373004357,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004325056060215265,
      "loss": 5.2787,
      "mean_token_accuracy": 0.17980556339025497,
      "num_tokens": 59455037.0,
      "step": 28655
    },
    {
      "entropy": 5.6210565090179445,
      "epoch": 2.5997822931785195,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004324824980775973,
      "loss": 5.3003,
      "mean_token_accuracy": 0.1826525405049324,
      "num_tokens": 59465093.0,
      "step": 28660
    },
    {
      "entropy": 5.6424788475036625,
      "epoch": 2.6002358490566038,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043245938687688995,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1879189968109131,
      "num_tokens": 59475449.0,
      "step": 28665
    },
    {
      "entropy": 5.713107299804688,
      "epoch": 2.600689404934688,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043243627241988235,
      "loss": 5.3123,
      "mean_token_accuracy": 0.185121950507164,
      "num_tokens": 59484990.0,
      "step": 28670
    },
    {
      "entropy": 5.668300676345825,
      "epoch": 2.6011429608127723,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043241315470705254,
      "loss": 5.3007,
      "mean_token_accuracy": 0.1795775130391121,
      "num_tokens": 59495739.0,
      "step": 28675
    },
    {
      "entropy": 5.59940276145935,
      "epoch": 2.601596516690856,
      "grad_norm": 2.5,
      "learning_rate": 0.00043239003373887855,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1841755360364914,
      "num_tokens": 59506432.0,
      "step": 28680
    },
    {
      "entropy": 5.658192539215088,
      "epoch": 2.6020500725689404,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004323669095158385,
      "loss": 5.2874,
      "mean_token_accuracy": 0.17958065420389174,
      "num_tokens": 59516585.0,
      "step": 28685
    },
    {
      "entropy": 5.618226003646851,
      "epoch": 2.6025036284470247,
      "grad_norm": 1.375,
      "learning_rate": 0.0004323437820384107,
      "loss": 5.3167,
      "mean_token_accuracy": 0.17995274513959886,
      "num_tokens": 59526824.0,
      "step": 28690
    },
    {
      "entropy": 5.7710175037384035,
      "epoch": 2.602957184325109,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004323206513070734,
      "loss": 5.3648,
      "mean_token_accuracy": 0.17554812878370285,
      "num_tokens": 59537749.0,
      "step": 28695
    },
    {
      "entropy": 5.650724458694458,
      "epoch": 2.603410740203193,
      "grad_norm": 2.203125,
      "learning_rate": 0.000432297517322305,
      "loss": 5.2935,
      "mean_token_accuracy": 0.17710775882005692,
      "num_tokens": 59548310.0,
      "step": 28700
    },
    {
      "entropy": 5.5985921859741214,
      "epoch": 2.6038642960812775,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004322743800845838,
      "loss": 5.3856,
      "mean_token_accuracy": 0.18255830258131028,
      "num_tokens": 59558191.0,
      "step": 28705
    },
    {
      "entropy": 5.709875869750976,
      "epoch": 2.6043178519593613,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043225123959438836,
      "loss": 5.3503,
      "mean_token_accuracy": 0.1743080958724022,
      "num_tokens": 59568961.0,
      "step": 28710
    },
    {
      "entropy": 5.64785943031311,
      "epoch": 2.6047714078374455,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043222809585219716,
      "loss": 5.3001,
      "mean_token_accuracy": 0.17907872200012206,
      "num_tokens": 59579137.0,
      "step": 28715
    },
    {
      "entropy": 5.4819587707519535,
      "epoch": 2.60522496371553,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043220494885848896,
      "loss": 5.1018,
      "mean_token_accuracy": 0.18960095643997193,
      "num_tokens": 59589612.0,
      "step": 28720
    },
    {
      "entropy": 5.704762744903564,
      "epoch": 2.605678519593614,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004321817986137423,
      "loss": 5.3439,
      "mean_token_accuracy": 0.17652182281017303,
      "num_tokens": 59600332.0,
      "step": 28725
    },
    {
      "entropy": 5.535020542144776,
      "epoch": 2.606132075471698,
      "grad_norm": 2.53125,
      "learning_rate": 0.000432158645118436,
      "loss": 5.2157,
      "mean_token_accuracy": 0.1945195198059082,
      "num_tokens": 59610491.0,
      "step": 28730
    },
    {
      "entropy": 5.7678844928741455,
      "epoch": 2.606585631349782,
      "grad_norm": 2.75,
      "learning_rate": 0.0004321354883730488,
      "loss": 5.4701,
      "mean_token_accuracy": 0.16778693944215775,
      "num_tokens": 59620729.0,
      "step": 28735
    },
    {
      "entropy": 5.664749240875244,
      "epoch": 2.6070391872278664,
      "grad_norm": 2.34375,
      "learning_rate": 0.00043211232837805973,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18595829457044602,
      "num_tokens": 59631243.0,
      "step": 28740
    },
    {
      "entropy": 5.60747013092041,
      "epoch": 2.6074927431059507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043208916513394767,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1848372235894203,
      "num_tokens": 59642055.0,
      "step": 28745
    },
    {
      "entropy": 5.674735355377197,
      "epoch": 2.607946298984035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043206599864119164,
      "loss": 5.3714,
      "mean_token_accuracy": 0.17890590578317642,
      "num_tokens": 59652144.0,
      "step": 28750
    },
    {
      "entropy": 5.692525196075439,
      "epoch": 2.6083998548621192,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00043204282890027074,
      "loss": 5.2541,
      "mean_token_accuracy": 0.19024807810783387,
      "num_tokens": 59662367.0,
      "step": 28755
    },
    {
      "entropy": 5.685465478897095,
      "epoch": 2.608853410740203,
      "grad_norm": 2.015625,
      "learning_rate": 0.00043201965591166413,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1794722080230713,
      "num_tokens": 59672729.0,
      "step": 28760
    },
    {
      "entropy": 5.705513525009155,
      "epoch": 2.6093069666182873,
      "grad_norm": 1.203125,
      "learning_rate": 0.000431996479675851,
      "loss": 5.3365,
      "mean_token_accuracy": 0.18326398730278015,
      "num_tokens": 59683282.0,
      "step": 28765
    },
    {
      "entropy": 5.667432641983032,
      "epoch": 2.6097605224963716,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043197330019331074,
      "loss": 5.4035,
      "mean_token_accuracy": 0.17715579271316528,
      "num_tokens": 59694297.0,
      "step": 28770
    },
    {
      "entropy": 5.569974136352539,
      "epoch": 2.610214078374456,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043195011746452244,
      "loss": 5.2337,
      "mean_token_accuracy": 0.18836402595043183,
      "num_tokens": 59704149.0,
      "step": 28775
    },
    {
      "entropy": 5.696934843063355,
      "epoch": 2.6106676342525397,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043192693148996585,
      "loss": 5.3602,
      "mean_token_accuracy": 0.1854990765452385,
      "num_tokens": 59714811.0,
      "step": 28780
    },
    {
      "entropy": 5.613439464569092,
      "epoch": 2.611121190130624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004319037422701203,
      "loss": 5.2436,
      "mean_token_accuracy": 0.18956104218959807,
      "num_tokens": 59724950.0,
      "step": 28785
    },
    {
      "entropy": 5.606035041809082,
      "epoch": 2.611574746008708,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004318805498054654,
      "loss": 5.3849,
      "mean_token_accuracy": 0.17216349989175797,
      "num_tokens": 59735905.0,
      "step": 28790
    },
    {
      "entropy": 5.720446491241455,
      "epoch": 2.6120283018867925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043185735409648077,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1799731358885765,
      "num_tokens": 59746743.0,
      "step": 28795
    },
    {
      "entropy": 5.692333841323853,
      "epoch": 2.6124818577648767,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043183415514364606,
      "loss": 5.2924,
      "mean_token_accuracy": 0.18293484598398208,
      "num_tokens": 59757205.0,
      "step": 28800
    },
    {
      "entropy": 5.601626443862915,
      "epoch": 2.612935413642961,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004318109529474411,
      "loss": 5.2787,
      "mean_token_accuracy": 0.18351584523916245,
      "num_tokens": 59767691.0,
      "step": 28805
    },
    {
      "entropy": 5.6044846534729,
      "epoch": 2.6133889695210453,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004317877475083456,
      "loss": 5.3094,
      "mean_token_accuracy": 0.1836305946111679,
      "num_tokens": 59778618.0,
      "step": 28810
    },
    {
      "entropy": 5.671081447601319,
      "epoch": 2.613842525399129,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043176453882683955,
      "loss": 5.3425,
      "mean_token_accuracy": 0.17943063676357268,
      "num_tokens": 59789423.0,
      "step": 28815
    },
    {
      "entropy": 5.715634870529175,
      "epoch": 2.6142960812772134,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043174132690340287,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1698588639497757,
      "num_tokens": 59800021.0,
      "step": 28820
    },
    {
      "entropy": 5.589719533920288,
      "epoch": 2.6147496371552976,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004317181117385156,
      "loss": 5.259,
      "mean_token_accuracy": 0.17476412355899812,
      "num_tokens": 59810006.0,
      "step": 28825
    },
    {
      "entropy": 5.662263011932373,
      "epoch": 2.6152031930333814,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043169489333265795,
      "loss": 5.3161,
      "mean_token_accuracy": 0.17735696882009505,
      "num_tokens": 59819505.0,
      "step": 28830
    },
    {
      "entropy": 5.712242603302002,
      "epoch": 2.6156567489114657,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043167167168630977,
      "loss": 5.3911,
      "mean_token_accuracy": 0.17567339390516282,
      "num_tokens": 59830439.0,
      "step": 28835
    },
    {
      "entropy": 5.639013862609863,
      "epoch": 2.61611030478955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004316484467999516,
      "loss": 5.3015,
      "mean_token_accuracy": 0.1865646496415138,
      "num_tokens": 59841281.0,
      "step": 28840
    },
    {
      "entropy": 5.655981779098511,
      "epoch": 2.6165638606676342,
      "grad_norm": 2.265625,
      "learning_rate": 0.00043162521867406376,
      "loss": 5.367,
      "mean_token_accuracy": 0.17064421325922013,
      "num_tokens": 59853581.0,
      "step": 28845
    },
    {
      "entropy": 5.753559160232544,
      "epoch": 2.6170174165457185,
      "grad_norm": 2.125,
      "learning_rate": 0.0004316019873091263,
      "loss": 5.3523,
      "mean_token_accuracy": 0.1821115329861641,
      "num_tokens": 59864787.0,
      "step": 28850
    },
    {
      "entropy": 5.6460987567901615,
      "epoch": 2.6174709724238028,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004315787527056199,
      "loss": 5.3246,
      "mean_token_accuracy": 0.18493559807538987,
      "num_tokens": 59875244.0,
      "step": 28855
    },
    {
      "entropy": 5.639953231811523,
      "epoch": 2.617924528301887,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004315555148640251,
      "loss": 5.3204,
      "mean_token_accuracy": 0.17631446719169616,
      "num_tokens": 59886535.0,
      "step": 28860
    },
    {
      "entropy": 5.688987159729004,
      "epoch": 2.618378084179971,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004315322737848222,
      "loss": 5.2723,
      "mean_token_accuracy": 0.1805923953652382,
      "num_tokens": 59896298.0,
      "step": 28865
    },
    {
      "entropy": 5.584215402603149,
      "epoch": 2.618831640058055,
      "grad_norm": 1.984375,
      "learning_rate": 0.000431509029468492,
      "loss": 5.2607,
      "mean_token_accuracy": 0.1866451695561409,
      "num_tokens": 59907010.0,
      "step": 28870
    },
    {
      "entropy": 5.69153938293457,
      "epoch": 2.6192851959361394,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004314857819155153,
      "loss": 5.2715,
      "mean_token_accuracy": 0.1897246465086937,
      "num_tokens": 59916247.0,
      "step": 28875
    },
    {
      "entropy": 5.607320356369018,
      "epoch": 2.6197387518142237,
      "grad_norm": 3.03125,
      "learning_rate": 0.00043146253112637267,
      "loss": 5.2062,
      "mean_token_accuracy": 0.19032161831855773,
      "num_tokens": 59926680.0,
      "step": 28880
    },
    {
      "entropy": 5.596458482742309,
      "epoch": 2.6201923076923075,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043143927710154504,
      "loss": 5.2104,
      "mean_token_accuracy": 0.19280908852815629,
      "num_tokens": 59936346.0,
      "step": 28885
    },
    {
      "entropy": 5.6489396572113035,
      "epoch": 2.6206458635703918,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004314160198415133,
      "loss": 5.3221,
      "mean_token_accuracy": 0.18439316600561143,
      "num_tokens": 59947007.0,
      "step": 28890
    },
    {
      "entropy": 5.682366228103637,
      "epoch": 2.621099419448476,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00043139275934675844,
      "loss": 5.358,
      "mean_token_accuracy": 0.18111313730478287,
      "num_tokens": 59957123.0,
      "step": 28895
    },
    {
      "entropy": 5.6253540992736815,
      "epoch": 2.6215529753265603,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004313694956177613,
      "loss": 5.3045,
      "mean_token_accuracy": 0.1862483486533165,
      "num_tokens": 59966566.0,
      "step": 28900
    },
    {
      "entropy": 5.637208747863769,
      "epoch": 2.6220065312046446,
      "grad_norm": 1.921875,
      "learning_rate": 0.00043134622865500334,
      "loss": 5.2609,
      "mean_token_accuracy": 0.17875730097293854,
      "num_tokens": 59976065.0,
      "step": 28905
    },
    {
      "entropy": 5.68057165145874,
      "epoch": 2.622460087082729,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004313229584589655,
      "loss": 5.2762,
      "mean_token_accuracy": 0.18542413264513016,
      "num_tokens": 59986201.0,
      "step": 28910
    },
    {
      "entropy": 5.5985692024230955,
      "epoch": 2.6229136429608126,
      "grad_norm": 1.703125,
      "learning_rate": 0.000431299685030129,
      "loss": 5.3104,
      "mean_token_accuracy": 0.1776291936635971,
      "num_tokens": 59996574.0,
      "step": 28915
    },
    {
      "entropy": 5.636337041854858,
      "epoch": 2.623367198838897,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00043127640836897516,
      "loss": 5.322,
      "mean_token_accuracy": 0.18202684968709945,
      "num_tokens": 60006082.0,
      "step": 28920
    },
    {
      "entropy": 5.624321937561035,
      "epoch": 2.623820754716981,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004312531284759854,
      "loss": 5.2943,
      "mean_token_accuracy": 0.18111149668693544,
      "num_tokens": 60015483.0,
      "step": 28925
    },
    {
      "entropy": 5.659129428863525,
      "epoch": 2.6242743105950654,
      "grad_norm": 2.09375,
      "learning_rate": 0.00043122984535164116,
      "loss": 5.24,
      "mean_token_accuracy": 0.19062529653310775,
      "num_tokens": 60025984.0,
      "step": 28930
    },
    {
      "entropy": 5.697665786743164,
      "epoch": 2.6247278664731493,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043120655899642386,
      "loss": 5.2919,
      "mean_token_accuracy": 0.1756989374756813,
      "num_tokens": 60036109.0,
      "step": 28935
    },
    {
      "entropy": 5.705066728591919,
      "epoch": 2.6251814223512335,
      "grad_norm": 3.515625,
      "learning_rate": 0.00043118326941081513,
      "loss": 5.3654,
      "mean_token_accuracy": 0.178032748401165,
      "num_tokens": 60046546.0,
      "step": 28940
    },
    {
      "entropy": 5.674505805969238,
      "epoch": 2.625634978229318,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004311599765952966,
      "loss": 5.3857,
      "mean_token_accuracy": 0.184739588201046,
      "num_tokens": 60056768.0,
      "step": 28945
    },
    {
      "entropy": 5.674889659881591,
      "epoch": 2.626088534107402,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043113668055034995,
      "loss": 5.4165,
      "mean_token_accuracy": 0.17949965894222258,
      "num_tokens": 60066962.0,
      "step": 28950
    },
    {
      "entropy": 5.651874351501465,
      "epoch": 2.6265420899854863,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004311133812764569,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18190246522426606,
      "num_tokens": 60076477.0,
      "step": 28955
    },
    {
      "entropy": 5.619127702713013,
      "epoch": 2.6269956458635706,
      "grad_norm": 3.234375,
      "learning_rate": 0.00043109007877409946,
      "loss": 5.196,
      "mean_token_accuracy": 0.18763407170772553,
      "num_tokens": 60088060.0,
      "step": 28960
    },
    {
      "entropy": 5.646897411346435,
      "epoch": 2.6274492017416544,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043106677304375935,
      "loss": 5.3271,
      "mean_token_accuracy": 0.1822088211774826,
      "num_tokens": 60098677.0,
      "step": 28965
    },
    {
      "entropy": 5.679339647293091,
      "epoch": 2.6279027576197387,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004310434640859186,
      "loss": 5.358,
      "mean_token_accuracy": 0.178207603096962,
      "num_tokens": 60109054.0,
      "step": 28970
    },
    {
      "entropy": 5.626439619064331,
      "epoch": 2.628356313497823,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004310201519010592,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18612127453088761,
      "num_tokens": 60119951.0,
      "step": 28975
    },
    {
      "entropy": 5.528684997558594,
      "epoch": 2.628809869375907,
      "grad_norm": 1.75,
      "learning_rate": 0.00043099683648966335,
      "loss": 5.1768,
      "mean_token_accuracy": 0.18479229211807252,
      "num_tokens": 60130026.0,
      "step": 28980
    },
    {
      "entropy": 5.573036432266235,
      "epoch": 2.629263425253991,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004309735178522132,
      "loss": 5.2058,
      "mean_token_accuracy": 0.19214953631162643,
      "num_tokens": 60140748.0,
      "step": 28985
    },
    {
      "entropy": 5.649463558197022,
      "epoch": 2.6297169811320753,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004309501959891909,
      "loss": 5.2691,
      "mean_token_accuracy": 0.1882981464266777,
      "num_tokens": 60151042.0,
      "step": 28990
    },
    {
      "entropy": 5.669949674606324,
      "epoch": 2.6301705370101596,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043092687090107883,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18072655647993088,
      "num_tokens": 60160758.0,
      "step": 28995
    },
    {
      "entropy": 5.570712995529175,
      "epoch": 2.630624092888244,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043090354258835934,
      "loss": 5.2947,
      "mean_token_accuracy": 0.17927112877368928,
      "num_tokens": 60170249.0,
      "step": 29000
    },
    {
      "entropy": 5.631701850891114,
      "epoch": 2.631077648766328,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043088021105151483,
      "loss": 5.3154,
      "mean_token_accuracy": 0.18180370479822158,
      "num_tokens": 60181216.0,
      "step": 29005
    },
    {
      "entropy": 5.665244388580322,
      "epoch": 2.6315312046444124,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004308568762910279,
      "loss": 5.2419,
      "mean_token_accuracy": 0.18660749197006227,
      "num_tokens": 60190559.0,
      "step": 29010
    },
    {
      "entropy": 5.7134472846984865,
      "epoch": 2.6319847605224966,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043083353830738097,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17589359879493713,
      "num_tokens": 60201268.0,
      "step": 29015
    },
    {
      "entropy": 5.542849731445313,
      "epoch": 2.6324383164005805,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043081019710105683,
      "loss": 5.126,
      "mean_token_accuracy": 0.19042403101921082,
      "num_tokens": 60210907.0,
      "step": 29020
    },
    {
      "entropy": 5.555708312988282,
      "epoch": 2.6328918722786647,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004307868526725381,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18937494307756425,
      "num_tokens": 60220298.0,
      "step": 29025
    },
    {
      "entropy": 5.60653510093689,
      "epoch": 2.633345428156749,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004307635050223075,
      "loss": 5.2337,
      "mean_token_accuracy": 0.19356247633695603,
      "num_tokens": 60230811.0,
      "step": 29030
    },
    {
      "entropy": 5.6590576171875,
      "epoch": 2.6337989840348333,
      "grad_norm": 1.875,
      "learning_rate": 0.000430740154150848,
      "loss": 5.2121,
      "mean_token_accuracy": 0.18531264066696168,
      "num_tokens": 60241211.0,
      "step": 29035
    },
    {
      "entropy": 5.606540203094482,
      "epoch": 2.634252539912917,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004307168000586423,
      "loss": 5.2644,
      "mean_token_accuracy": 0.18452896922826767,
      "num_tokens": 60251506.0,
      "step": 29040
    },
    {
      "entropy": 5.59657506942749,
      "epoch": 2.6347060957910013,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004306934427461736,
      "loss": 5.255,
      "mean_token_accuracy": 0.19348880499601365,
      "num_tokens": 60261416.0,
      "step": 29045
    },
    {
      "entropy": 5.597267007827758,
      "epoch": 2.6351596516690856,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004306700822139248,
      "loss": 5.2901,
      "mean_token_accuracy": 0.17998429238796235,
      "num_tokens": 60272315.0,
      "step": 29050
    },
    {
      "entropy": 5.643957662582397,
      "epoch": 2.63561320754717,
      "grad_norm": 1.515625,
      "learning_rate": 0.000430646718462379,
      "loss": 5.3671,
      "mean_token_accuracy": 0.18180346637964248,
      "num_tokens": 60282722.0,
      "step": 29055
    },
    {
      "entropy": 5.631772089004516,
      "epoch": 2.636066763425254,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004306233514920194,
      "loss": 5.2897,
      "mean_token_accuracy": 0.18333445489406586,
      "num_tokens": 60292636.0,
      "step": 29060
    },
    {
      "entropy": 5.603409719467163,
      "epoch": 2.6365203193033384,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00043059998130332933,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18643279522657394,
      "num_tokens": 60302979.0,
      "step": 29065
    },
    {
      "entropy": 5.685333919525147,
      "epoch": 2.6369738751814222,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043057660789679194,
      "loss": 5.2989,
      "mean_token_accuracy": 0.1835133895277977,
      "num_tokens": 60313181.0,
      "step": 29070
    },
    {
      "entropy": 5.6866086483001705,
      "epoch": 2.6374274310595065,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004305532312728906,
      "loss": 5.232,
      "mean_token_accuracy": 0.18195315152406694,
      "num_tokens": 60323401.0,
      "step": 29075
    },
    {
      "entropy": 5.615885829925537,
      "epoch": 2.6378809869375908,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004305298514321088,
      "loss": 5.3421,
      "mean_token_accuracy": 0.18360238373279572,
      "num_tokens": 60333478.0,
      "step": 29080
    },
    {
      "entropy": 5.6559051990509035,
      "epoch": 2.638334542815675,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043050646837493014,
      "loss": 5.2519,
      "mean_token_accuracy": 0.18510946929454802,
      "num_tokens": 60343980.0,
      "step": 29085
    },
    {
      "entropy": 5.634364795684815,
      "epoch": 2.638788098693759,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004304830821018381,
      "loss": 5.2693,
      "mean_token_accuracy": 0.18495581299066544,
      "num_tokens": 60354262.0,
      "step": 29090
    },
    {
      "entropy": 5.647772645950317,
      "epoch": 2.639241654571843,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004304596926133162,
      "loss": 5.3338,
      "mean_token_accuracy": 0.18281057626008987,
      "num_tokens": 60364137.0,
      "step": 29095
    },
    {
      "entropy": 5.624353075027466,
      "epoch": 2.6396952104499274,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043043629990984823,
      "loss": 5.3361,
      "mean_token_accuracy": 0.17660763412714003,
      "num_tokens": 60374528.0,
      "step": 29100
    },
    {
      "entropy": 5.660975933074951,
      "epoch": 2.6401487663280117,
      "grad_norm": 2.375,
      "learning_rate": 0.00043041290399191797,
      "loss": 5.2931,
      "mean_token_accuracy": 0.18938452005386353,
      "num_tokens": 60384880.0,
      "step": 29105
    },
    {
      "entropy": 5.653546905517578,
      "epoch": 2.640602322206096,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004303895048600093,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18041754513978958,
      "num_tokens": 60395543.0,
      "step": 29110
    },
    {
      "entropy": 5.6421855926513675,
      "epoch": 2.64105587808418,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004303661025146061,
      "loss": 5.2643,
      "mean_token_accuracy": 0.18556291460990906,
      "num_tokens": 60405432.0,
      "step": 29115
    },
    {
      "entropy": 5.663226890563965,
      "epoch": 2.641509433962264,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004303426969561922,
      "loss": 5.3519,
      "mean_token_accuracy": 0.174553582072258,
      "num_tokens": 60416360.0,
      "step": 29120
    },
    {
      "entropy": 5.642409324645996,
      "epoch": 2.6419629898403483,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004303192881852518,
      "loss": 5.3002,
      "mean_token_accuracy": 0.18131573796272277,
      "num_tokens": 60426371.0,
      "step": 29125
    },
    {
      "entropy": 5.649956321716308,
      "epoch": 2.6424165457184325,
      "grad_norm": 2.0625,
      "learning_rate": 0.00043029587620226897,
      "loss": 5.3112,
      "mean_token_accuracy": 0.17729413658380508,
      "num_tokens": 60437505.0,
      "step": 29130
    },
    {
      "entropy": 5.674519634246826,
      "epoch": 2.642870101596517,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004302724610077278,
      "loss": 5.3038,
      "mean_token_accuracy": 0.18083688169717788,
      "num_tokens": 60448170.0,
      "step": 29135
    },
    {
      "entropy": 5.685340023040771,
      "epoch": 2.6433236574746006,
      "grad_norm": 1.625,
      "learning_rate": 0.0004302490426021126,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17502436190843582,
      "num_tokens": 60458776.0,
      "step": 29140
    },
    {
      "entropy": 5.646800231933594,
      "epoch": 2.643777213352685,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004302256209859076,
      "loss": 5.2711,
      "mean_token_accuracy": 0.1905123248696327,
      "num_tokens": 60468279.0,
      "step": 29145
    },
    {
      "entropy": 5.632421493530273,
      "epoch": 2.644230769230769,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043020219615959727,
      "loss": 5.351,
      "mean_token_accuracy": 0.17933713942766188,
      "num_tokens": 60478507.0,
      "step": 29150
    },
    {
      "entropy": 5.757854986190796,
      "epoch": 2.6446843251088534,
      "grad_norm": 1.25,
      "learning_rate": 0.0004301787681236659,
      "loss": 5.3959,
      "mean_token_accuracy": 0.17369102090597152,
      "num_tokens": 60489016.0,
      "step": 29155
    },
    {
      "entropy": 5.756720066070557,
      "epoch": 2.6451378809869377,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004301553368785981,
      "loss": 5.3567,
      "mean_token_accuracy": 0.1735457345843315,
      "num_tokens": 60500169.0,
      "step": 29160
    },
    {
      "entropy": 5.5929821014404295,
      "epoch": 2.645591436865022,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004301319024248784,
      "loss": 5.2637,
      "mean_token_accuracy": 0.1836981400847435,
      "num_tokens": 60511109.0,
      "step": 29165
    },
    {
      "entropy": 5.629164505004883,
      "epoch": 2.6460449927431062,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043010846476299145,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18785209506750106,
      "num_tokens": 60521735.0,
      "step": 29170
    },
    {
      "entropy": 5.72378306388855,
      "epoch": 2.64649854862119,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004300850238934219,
      "loss": 5.3304,
      "mean_token_accuracy": 0.17977936267852784,
      "num_tokens": 60532071.0,
      "step": 29175
    },
    {
      "entropy": 5.671810674667358,
      "epoch": 2.6469521044992743,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00043006157981665446,
      "loss": 5.2908,
      "mean_token_accuracy": 0.18239894956350328,
      "num_tokens": 60541909.0,
      "step": 29180
    },
    {
      "entropy": 5.539666414260864,
      "epoch": 2.6474056603773586,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043003813253317415,
      "loss": 5.2394,
      "mean_token_accuracy": 0.19054187834262848,
      "num_tokens": 60551361.0,
      "step": 29185
    },
    {
      "entropy": 5.611055088043213,
      "epoch": 2.6478592162554424,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043001468204346564,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1793530136346817,
      "num_tokens": 60562112.0,
      "step": 29190
    },
    {
      "entropy": 5.682084655761718,
      "epoch": 2.6483127721335267,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00042999122834801406,
      "loss": 5.3264,
      "mean_token_accuracy": 0.17816998511552812,
      "num_tokens": 60572827.0,
      "step": 29195
    },
    {
      "entropy": 5.563035202026367,
      "epoch": 2.648766328011611,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004299677714473044,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18633523881435393,
      "num_tokens": 60582222.0,
      "step": 29200
    },
    {
      "entropy": 5.664576721191406,
      "epoch": 2.649219883889695,
      "grad_norm": 2.125,
      "learning_rate": 0.0004299443113418217,
      "loss": 5.3324,
      "mean_token_accuracy": 0.18650047183036805,
      "num_tokens": 60592943.0,
      "step": 29205
    },
    {
      "entropy": 5.670470523834228,
      "epoch": 2.6496734397677795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042992084803205124,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17683335542678832,
      "num_tokens": 60603206.0,
      "step": 29210
    },
    {
      "entropy": 5.571117782592774,
      "epoch": 2.6501269956458637,
      "grad_norm": 1.625,
      "learning_rate": 0.000429897381518478,
      "loss": 5.2532,
      "mean_token_accuracy": 0.18601265996694566,
      "num_tokens": 60614776.0,
      "step": 29215
    },
    {
      "entropy": 5.634550619125366,
      "epoch": 2.650580551523948,
      "grad_norm": 9.625,
      "learning_rate": 0.00042987391180158757,
      "loss": 5.2543,
      "mean_token_accuracy": 0.17877203971147537,
      "num_tokens": 60625914.0,
      "step": 29220
    },
    {
      "entropy": 5.653992795944214,
      "epoch": 2.651034107402032,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042985043888186516,
      "loss": 5.2879,
      "mean_token_accuracy": 0.1869732543826103,
      "num_tokens": 60636934.0,
      "step": 29225
    },
    {
      "entropy": 5.632002544403076,
      "epoch": 2.651487663280116,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004298269627597962,
      "loss": 5.3138,
      "mean_token_accuracy": 0.1776123970746994,
      "num_tokens": 60647707.0,
      "step": 29230
    },
    {
      "entropy": 5.524977445602417,
      "epoch": 2.6519412191582004,
      "grad_norm": 1.84375,
      "learning_rate": 0.00042980348343586624,
      "loss": 5.2471,
      "mean_token_accuracy": 0.18053716719150542,
      "num_tokens": 60658691.0,
      "step": 29235
    },
    {
      "entropy": 5.655204200744629,
      "epoch": 2.6523947750362846,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00042978000091056074,
      "loss": 5.3376,
      "mean_token_accuracy": 0.18105473071336747,
      "num_tokens": 60668479.0,
      "step": 29240
    },
    {
      "entropy": 5.644942426681519,
      "epoch": 2.6528483309143684,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004297565151843653,
      "loss": 5.2016,
      "mean_token_accuracy": 0.18865847885608672,
      "num_tokens": 60679158.0,
      "step": 29245
    },
    {
      "entropy": 5.695441627502442,
      "epoch": 2.6533018867924527,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004297330262577659,
      "loss": 5.4022,
      "mean_token_accuracy": 0.17828623652458192,
      "num_tokens": 60689363.0,
      "step": 29250
    },
    {
      "entropy": 5.596778345108032,
      "epoch": 2.653755442670537,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004297095341312479,
      "loss": 5.1666,
      "mean_token_accuracy": 0.18582468330860138,
      "num_tokens": 60699650.0,
      "step": 29255
    },
    {
      "entropy": 5.584618377685547,
      "epoch": 2.6542089985486212,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004296860388052974,
      "loss": 5.267,
      "mean_token_accuracy": 0.18279835134744643,
      "num_tokens": 60709921.0,
      "step": 29260
    },
    {
      "entropy": 5.6738239288330075,
      "epoch": 2.6546625544267055,
      "grad_norm": 1.625,
      "learning_rate": 0.00042966254028040015,
      "loss": 5.261,
      "mean_token_accuracy": 0.1872641235589981,
      "num_tokens": 60720112.0,
      "step": 29265
    },
    {
      "entropy": 5.62284779548645,
      "epoch": 2.65511611030479,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042963903855704216,
      "loss": 5.1938,
      "mean_token_accuracy": 0.18094307631254197,
      "num_tokens": 60730604.0,
      "step": 29270
    },
    {
      "entropy": 5.571313714981079,
      "epoch": 2.6555696661828736,
      "grad_norm": 1.859375,
      "learning_rate": 0.00042961553363570946,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18476882129907607,
      "num_tokens": 60740896.0,
      "step": 29275
    },
    {
      "entropy": 5.635297536849976,
      "epoch": 2.656023222060958,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004295920255168881,
      "loss": 5.2685,
      "mean_token_accuracy": 0.1840734750032425,
      "num_tokens": 60751771.0,
      "step": 29280
    },
    {
      "entropy": 5.697774934768677,
      "epoch": 2.656476777939042,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00042956851420106423,
      "loss": 5.3127,
      "mean_token_accuracy": 0.18111418932676315,
      "num_tokens": 60762802.0,
      "step": 29285
    },
    {
      "entropy": 5.577886724472046,
      "epoch": 2.6569303338171264,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004295449996887242,
      "loss": 5.2184,
      "mean_token_accuracy": 0.19070041924715042,
      "num_tokens": 60774102.0,
      "step": 29290
    },
    {
      "entropy": 5.627483463287353,
      "epoch": 2.65738388969521,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004295214819803541,
      "loss": 5.2892,
      "mean_token_accuracy": 0.18201993256807328,
      "num_tokens": 60785383.0,
      "step": 29295
    },
    {
      "entropy": 5.609569072723389,
      "epoch": 2.6578374455732945,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00042949796107644053,
      "loss": 5.2815,
      "mean_token_accuracy": 0.18916800767183303,
      "num_tokens": 60795173.0,
      "step": 29300
    },
    {
      "entropy": 5.659349346160889,
      "epoch": 2.6582910014513788,
      "grad_norm": 2.03125,
      "learning_rate": 0.00042947443697746956,
      "loss": 5.2766,
      "mean_token_accuracy": 0.18937930911779405,
      "num_tokens": 60805621.0,
      "step": 29305
    },
    {
      "entropy": 5.711781167984009,
      "epoch": 2.658744557329463,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004294509096839279,
      "loss": 5.4152,
      "mean_token_accuracy": 0.17021520882844926,
      "num_tokens": 60816799.0,
      "step": 29310
    },
    {
      "entropy": 5.589362907409668,
      "epoch": 2.6591981132075473,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042942737919630215,
      "loss": 5.2182,
      "mean_token_accuracy": 0.1858784332871437,
      "num_tokens": 60826827.0,
      "step": 29315
    },
    {
      "entropy": 5.631322860717773,
      "epoch": 2.6596516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042940384551507876,
      "loss": 5.281,
      "mean_token_accuracy": 0.1834815338253975,
      "num_tokens": 60836567.0,
      "step": 29320
    },
    {
      "entropy": 5.633705472946167,
      "epoch": 2.6601052249637154,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004293803086407447,
      "loss": 5.2976,
      "mean_token_accuracy": 0.17682288587093353,
      "num_tokens": 60846723.0,
      "step": 29325
    },
    {
      "entropy": 5.650498247146606,
      "epoch": 2.6605587808417996,
      "grad_norm": 1.375,
      "learning_rate": 0.0004293567685737863,
      "loss": 5.3861,
      "mean_token_accuracy": 0.17837442755699157,
      "num_tokens": 60857434.0,
      "step": 29330
    },
    {
      "entropy": 5.69902663230896,
      "epoch": 2.661012336719884,
      "grad_norm": 1.734375,
      "learning_rate": 0.00042933322531469074,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1826843649148941,
      "num_tokens": 60868780.0,
      "step": 29335
    },
    {
      "entropy": 5.704816150665283,
      "epoch": 2.661465892597968,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042930967886394456,
      "loss": 5.3487,
      "mean_token_accuracy": 0.18769324272871019,
      "num_tokens": 60879037.0,
      "step": 29340
    },
    {
      "entropy": 5.566553592681885,
      "epoch": 2.661919448476052,
      "grad_norm": 1.5,
      "learning_rate": 0.000429286129222035,
      "loss": 5.2574,
      "mean_token_accuracy": 0.1865200400352478,
      "num_tokens": 60889543.0,
      "step": 29345
    },
    {
      "entropy": 5.563624858856201,
      "epoch": 2.6623730043541363,
      "grad_norm": 2.921875,
      "learning_rate": 0.000429262576389449,
      "loss": 5.2755,
      "mean_token_accuracy": 0.18260575830936432,
      "num_tokens": 60900141.0,
      "step": 29350
    },
    {
      "entropy": 5.662040328979492,
      "epoch": 2.6628265602322205,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004292390203666736,
      "loss": 5.2921,
      "mean_token_accuracy": 0.1896065354347229,
      "num_tokens": 60910244.0,
      "step": 29355
    },
    {
      "entropy": 5.628397750854492,
      "epoch": 2.663280116110305,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000429215461154196,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1939480111002922,
      "num_tokens": 60919908.0,
      "step": 29360
    },
    {
      "entropy": 5.60447359085083,
      "epoch": 2.663733671988389,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042919189875250324,
      "loss": 5.2775,
      "mean_token_accuracy": 0.18230955302715302,
      "num_tokens": 60929096.0,
      "step": 29365
    },
    {
      "entropy": 5.5977448463439945,
      "epoch": 2.6641872278664733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042916833316208284,
      "loss": 5.2759,
      "mean_token_accuracy": 0.17926269471645356,
      "num_tokens": 60939346.0,
      "step": 29370
    },
    {
      "entropy": 5.688455247879029,
      "epoch": 2.6646407837445576,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004291447643834219,
      "loss": 5.2995,
      "mean_token_accuracy": 0.17965984642505645,
      "num_tokens": 60950364.0,
      "step": 29375
    },
    {
      "entropy": 5.665002155303955,
      "epoch": 2.6650943396226414,
      "grad_norm": 1.9375,
      "learning_rate": 0.000429121192417008,
      "loss": 5.2708,
      "mean_token_accuracy": 0.1857529103755951,
      "num_tokens": 60960905.0,
      "step": 29380
    },
    {
      "entropy": 5.685887289047241,
      "epoch": 2.6655478955007257,
      "grad_norm": 2.34375,
      "learning_rate": 0.00042909761726332864,
      "loss": 5.3184,
      "mean_token_accuracy": 0.18989420235157012,
      "num_tokens": 60971281.0,
      "step": 29385
    },
    {
      "entropy": 5.590190601348877,
      "epoch": 2.66600145137881,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004290740389228712,
      "loss": 5.3323,
      "mean_token_accuracy": 0.18147220760583876,
      "num_tokens": 60982522.0,
      "step": 29390
    },
    {
      "entropy": 5.605115270614624,
      "epoch": 2.6664550072568938,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00042905045739612347,
      "loss": 5.2944,
      "mean_token_accuracy": 0.1851483479142189,
      "num_tokens": 60992738.0,
      "step": 29395
    },
    {
      "entropy": 5.71400032043457,
      "epoch": 2.666908563134978,
      "grad_norm": 1.625,
      "learning_rate": 0.00042902687268357297,
      "loss": 5.4167,
      "mean_token_accuracy": 0.18241459876298904,
      "num_tokens": 61003665.0,
      "step": 29400
    },
    {
      "entropy": 5.816492128372192,
      "epoch": 2.6673621190130623,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042900328478570745,
      "loss": 5.4325,
      "mean_token_accuracy": 0.1787679448723793,
      "num_tokens": 61014663.0,
      "step": 29405
    },
    {
      "entropy": 5.631447267532349,
      "epoch": 2.6678156748911466,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004289796937030148,
      "loss": 5.21,
      "mean_token_accuracy": 0.18876786530017853,
      "num_tokens": 61024916.0,
      "step": 29410
    },
    {
      "entropy": 5.546574831008911,
      "epoch": 2.668269230769231,
      "grad_norm": 1.625,
      "learning_rate": 0.00042895609943598286,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18560840636491777,
      "num_tokens": 61035043.0,
      "step": 29415
    },
    {
      "entropy": 5.633942604064941,
      "epoch": 2.668722786647315,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004289325019850995,
      "loss": 5.3505,
      "mean_token_accuracy": 0.17593094557523728,
      "num_tokens": 61046153.0,
      "step": 29420
    },
    {
      "entropy": 5.713750410079956,
      "epoch": 2.6691763425253994,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004289089013508528,
      "loss": 5.259,
      "mean_token_accuracy": 0.18422684520483018,
      "num_tokens": 61057009.0,
      "step": 29425
    },
    {
      "entropy": 5.680473899841308,
      "epoch": 2.669629898403483,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004288852975337308,
      "loss": 5.4088,
      "mean_token_accuracy": 0.17334595769643785,
      "num_tokens": 61068514.0,
      "step": 29430
    },
    {
      "entropy": 5.661332082748413,
      "epoch": 2.6700834542815675,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042886169053422165,
      "loss": 5.3083,
      "mean_token_accuracy": 0.18429206907749177,
      "num_tokens": 61078902.0,
      "step": 29435
    },
    {
      "entropy": 5.633921909332275,
      "epoch": 2.6705370101596517,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004288380803528135,
      "loss": 5.3056,
      "mean_token_accuracy": 0.19031433761119843,
      "num_tokens": 61089216.0,
      "step": 29440
    },
    {
      "entropy": 5.589647388458252,
      "epoch": 2.670990566037736,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042881446698999465,
      "loss": 5.2038,
      "mean_token_accuracy": 0.19513365477323533,
      "num_tokens": 61098470.0,
      "step": 29445
    },
    {
      "entropy": 5.739257717132569,
      "epoch": 2.67144412191582,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004287908504462534,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18716476261615753,
      "num_tokens": 61108694.0,
      "step": 29450
    },
    {
      "entropy": 5.638243532180786,
      "epoch": 2.671897677793904,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042876723072207825,
      "loss": 5.2681,
      "mean_token_accuracy": 0.1851534962654114,
      "num_tokens": 61119389.0,
      "step": 29455
    },
    {
      "entropy": 5.702775192260742,
      "epoch": 2.6723512336719883,
      "grad_norm": 1.5625,
      "learning_rate": 0.00042874360781795757,
      "loss": 5.3677,
      "mean_token_accuracy": 0.17933852672576905,
      "num_tokens": 61130491.0,
      "step": 29460
    },
    {
      "entropy": 5.631885671615601,
      "epoch": 2.6728047895500726,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00042871998173437987,
      "loss": 5.3148,
      "mean_token_accuracy": 0.1867539569735527,
      "num_tokens": 61140404.0,
      "step": 29465
    },
    {
      "entropy": 5.681228446960449,
      "epoch": 2.673258345428157,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042869635247183373,
      "loss": 5.3757,
      "mean_token_accuracy": 0.18097400218248366,
      "num_tokens": 61150777.0,
      "step": 29470
    },
    {
      "entropy": 5.630759048461914,
      "epoch": 2.673711901306241,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042867272003080793,
      "loss": 5.3037,
      "mean_token_accuracy": 0.18477047681808473,
      "num_tokens": 61160685.0,
      "step": 29475
    },
    {
      "entropy": 5.669190120697022,
      "epoch": 2.674165457184325,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004286490844117911,
      "loss": 5.2721,
      "mean_token_accuracy": 0.18133708834648132,
      "num_tokens": 61170553.0,
      "step": 29480
    },
    {
      "entropy": 5.722750902175903,
      "epoch": 2.6746190130624092,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042862544561527203,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17950277775526047,
      "num_tokens": 61181396.0,
      "step": 29485
    },
    {
      "entropy": 5.635095643997192,
      "epoch": 2.6750725689404935,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004286018036417396,
      "loss": 5.3248,
      "mean_token_accuracy": 0.18304179906845092,
      "num_tokens": 61192315.0,
      "step": 29490
    },
    {
      "entropy": 5.630108737945557,
      "epoch": 2.6755261248185778,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00042857815849168277,
      "loss": 5.2807,
      "mean_token_accuracy": 0.18601104170083999,
      "num_tokens": 61202719.0,
      "step": 29495
    },
    {
      "entropy": 5.548515272140503,
      "epoch": 2.6759796806966616,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004285545101655904,
      "loss": 5.1578,
      "mean_token_accuracy": 0.18774113357067107,
      "num_tokens": 61212703.0,
      "step": 29500
    },
    {
      "entropy": 5.5286438941955565,
      "epoch": 2.676433236574746,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004285308586639517,
      "loss": 5.1331,
      "mean_token_accuracy": 0.1949339345097542,
      "num_tokens": 61222406.0,
      "step": 29505
    },
    {
      "entropy": 5.617227792739868,
      "epoch": 2.67688679245283,
      "grad_norm": 1.375,
      "learning_rate": 0.00042850720398725575,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17346716821193695,
      "num_tokens": 61232758.0,
      "step": 29510
    },
    {
      "entropy": 5.647122859954834,
      "epoch": 2.6773403483309144,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042848354613599164,
      "loss": 5.2644,
      "mean_token_accuracy": 0.19257582277059554,
      "num_tokens": 61243715.0,
      "step": 29515
    },
    {
      "entropy": 5.656592130661011,
      "epoch": 2.6777939042089987,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042845988511064863,
      "loss": 5.3163,
      "mean_token_accuracy": 0.18301938325166703,
      "num_tokens": 61254669.0,
      "step": 29520
    },
    {
      "entropy": 5.65608229637146,
      "epoch": 2.678247460087083,
      "grad_norm": 2.46875,
      "learning_rate": 0.00042843622091171625,
      "loss": 5.287,
      "mean_token_accuracy": 0.1872505232691765,
      "num_tokens": 61264377.0,
      "step": 29525
    },
    {
      "entropy": 5.597886991500855,
      "epoch": 2.6787010159651667,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004284125535396836,
      "loss": 5.2898,
      "mean_token_accuracy": 0.18027803301811218,
      "num_tokens": 61275348.0,
      "step": 29530
    },
    {
      "entropy": 5.5108945846557615,
      "epoch": 2.679154571843251,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042838888299504024,
      "loss": 5.1358,
      "mean_token_accuracy": 0.19125176668167115,
      "num_tokens": 61285579.0,
      "step": 29535
    },
    {
      "entropy": 5.61525297164917,
      "epoch": 2.6796081277213353,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00042836520927827575,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1834685668349266,
      "num_tokens": 61296037.0,
      "step": 29540
    },
    {
      "entropy": 5.651497840881348,
      "epoch": 2.6800616835994195,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004283415323898797,
      "loss": 5.2267,
      "mean_token_accuracy": 0.19549481719732284,
      "num_tokens": 61305882.0,
      "step": 29545
    },
    {
      "entropy": 5.610578393936157,
      "epoch": 2.6805152394775034,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042831785233034156,
      "loss": 5.2217,
      "mean_token_accuracy": 0.18755552768707276,
      "num_tokens": 61317117.0,
      "step": 29550
    },
    {
      "entropy": 5.635826969146729,
      "epoch": 2.6809687953555876,
      "grad_norm": 2.0625,
      "learning_rate": 0.00042829416910015116,
      "loss": 5.3114,
      "mean_token_accuracy": 0.18099390864372253,
      "num_tokens": 61328190.0,
      "step": 29555
    },
    {
      "entropy": 5.6287164211273195,
      "epoch": 2.681422351233672,
      "grad_norm": 1.484375,
      "learning_rate": 0.00042827048269979834,
      "loss": 5.2981,
      "mean_token_accuracy": 0.18346291333436965,
      "num_tokens": 61337978.0,
      "step": 29560
    },
    {
      "entropy": 5.645231437683106,
      "epoch": 2.681875907111756,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004282467931297729,
      "loss": 5.3467,
      "mean_token_accuracy": 0.17973437905311584,
      "num_tokens": 61348523.0,
      "step": 29565
    },
    {
      "entropy": 5.619403457641601,
      "epoch": 2.6823294629898404,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004282231003905647,
      "loss": 5.2441,
      "mean_token_accuracy": 0.19029568135738373,
      "num_tokens": 61359250.0,
      "step": 29570
    },
    {
      "entropy": 5.613372373580932,
      "epoch": 2.6827830188679247,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00042819940448266376,
      "loss": 5.2005,
      "mean_token_accuracy": 0.1884540304541588,
      "num_tokens": 61369999.0,
      "step": 29575
    },
    {
      "entropy": 5.688187599182129,
      "epoch": 2.683236574746009,
      "grad_norm": 2.015625,
      "learning_rate": 0.00042817570540656003,
      "loss": 5.3229,
      "mean_token_accuracy": 0.18095456063747406,
      "num_tokens": 61380279.0,
      "step": 29580
    },
    {
      "entropy": 5.589051198959351,
      "epoch": 2.683690130624093,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004281520031627438,
      "loss": 5.2898,
      "mean_token_accuracy": 0.18887605667114257,
      "num_tokens": 61389794.0,
      "step": 29585
    },
    {
      "entropy": 5.653131484985352,
      "epoch": 2.684143686502177,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042812829775170497,
      "loss": 5.4252,
      "mean_token_accuracy": 0.17797132581472397,
      "num_tokens": 61401136.0,
      "step": 29590
    },
    {
      "entropy": 5.677339649200439,
      "epoch": 2.6845972423802613,
      "grad_norm": 2.25,
      "learning_rate": 0.00042810458917393404,
      "loss": 5.3476,
      "mean_token_accuracy": 0.17992305010557175,
      "num_tokens": 61412830.0,
      "step": 29595
    },
    {
      "entropy": 5.6285865783691404,
      "epoch": 2.6850507982583456,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004280808774299211,
      "loss": 5.2214,
      "mean_token_accuracy": 0.1921207770705223,
      "num_tokens": 61422168.0,
      "step": 29600
    },
    {
      "entropy": 5.539923429489136,
      "epoch": 2.6855043541364294,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042805716252015665,
      "loss": 5.2219,
      "mean_token_accuracy": 0.19034148305654525,
      "num_tokens": 61432471.0,
      "step": 29605
    },
    {
      "entropy": 5.623132514953613,
      "epoch": 2.6859579100145137,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042803344444513106,
      "loss": 5.2705,
      "mean_token_accuracy": 0.18717418760061263,
      "num_tokens": 61443402.0,
      "step": 29610
    },
    {
      "entropy": 5.7101904392242435,
      "epoch": 2.686411465892598,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004280097232053349,
      "loss": 5.2714,
      "mean_token_accuracy": 0.18135614693164825,
      "num_tokens": 61454465.0,
      "step": 29615
    },
    {
      "entropy": 5.6074547290802,
      "epoch": 2.686865021770682,
      "grad_norm": 2.25,
      "learning_rate": 0.0004279859988012586,
      "loss": 5.2818,
      "mean_token_accuracy": 0.1842081919312477,
      "num_tokens": 61464671.0,
      "step": 29620
    },
    {
      "entropy": 5.595145225524902,
      "epoch": 2.6873185776487665,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004279622712333929,
      "loss": 5.2401,
      "mean_token_accuracy": 0.18958383649587632,
      "num_tokens": 61474940.0,
      "step": 29625
    },
    {
      "entropy": 5.648260116577148,
      "epoch": 2.6877721335268507,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004279385405022284,
      "loss": 5.304,
      "mean_token_accuracy": 0.18192558288574218,
      "num_tokens": 61485240.0,
      "step": 29630
    },
    {
      "entropy": 5.606990098953247,
      "epoch": 2.6882256894049346,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000427914806608256,
      "loss": 5.2476,
      "mean_token_accuracy": 0.18434574753046035,
      "num_tokens": 61495382.0,
      "step": 29635
    },
    {
      "entropy": 5.558651447296143,
      "epoch": 2.688679245283019,
      "grad_norm": 3.375,
      "learning_rate": 0.0004278910695519663,
      "loss": 5.1955,
      "mean_token_accuracy": 0.19057289510965347,
      "num_tokens": 61505372.0,
      "step": 29640
    },
    {
      "entropy": 5.6645262241363525,
      "epoch": 2.689132801161103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042786732933385046,
      "loss": 5.2905,
      "mean_token_accuracy": 0.19075305312871932,
      "num_tokens": 61516039.0,
      "step": 29645
    },
    {
      "entropy": 5.6719433784484865,
      "epoch": 2.6895863570391874,
      "grad_norm": 1.875,
      "learning_rate": 0.0004278435859543992,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1842140480875969,
      "num_tokens": 61525435.0,
      "step": 29650
    },
    {
      "entropy": 5.616942024230957,
      "epoch": 2.690039912917271,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004278198394141037,
      "loss": 5.2684,
      "mean_token_accuracy": 0.17925487458705902,
      "num_tokens": 61534714.0,
      "step": 29655
    },
    {
      "entropy": 5.662509965896606,
      "epoch": 2.6904934687953554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042779608971345496,
      "loss": 5.4182,
      "mean_token_accuracy": 0.17567117363214493,
      "num_tokens": 61545841.0,
      "step": 29660
    },
    {
      "entropy": 5.616568326950073,
      "epoch": 2.6909470246734397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042777233685294407,
      "loss": 5.186,
      "mean_token_accuracy": 0.1953650251030922,
      "num_tokens": 61555355.0,
      "step": 29665
    },
    {
      "entropy": 5.65064377784729,
      "epoch": 2.691400580551524,
      "grad_norm": 1.375,
      "learning_rate": 0.00042774858083306246,
      "loss": 5.3244,
      "mean_token_accuracy": 0.17786937654018403,
      "num_tokens": 61565474.0,
      "step": 29670
    },
    {
      "entropy": 5.650510168075561,
      "epoch": 2.6918541364296082,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042772482165430117,
      "loss": 5.2114,
      "mean_token_accuracy": 0.19087353348731995,
      "num_tokens": 61575631.0,
      "step": 29675
    },
    {
      "entropy": 5.591071224212646,
      "epoch": 2.6923076923076925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042770105931715167,
      "loss": 5.2613,
      "mean_token_accuracy": 0.184549917280674,
      "num_tokens": 61586398.0,
      "step": 29680
    },
    {
      "entropy": 5.595301342010498,
      "epoch": 2.6927612481857763,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004276772938221054,
      "loss": 5.1943,
      "mean_token_accuracy": 0.19227590262889863,
      "num_tokens": 61596630.0,
      "step": 29685
    },
    {
      "entropy": 5.631232023239136,
      "epoch": 2.6932148040638606,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004276535251696537,
      "loss": 5.2953,
      "mean_token_accuracy": 0.18700500279664994,
      "num_tokens": 61605965.0,
      "step": 29690
    },
    {
      "entropy": 5.6970741748809814,
      "epoch": 2.693668359941945,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004276297533602883,
      "loss": 5.3685,
      "mean_token_accuracy": 0.18164975494146346,
      "num_tokens": 61616973.0,
      "step": 29695
    },
    {
      "entropy": 5.717730093002319,
      "epoch": 2.694121915820029,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042760597839450057,
      "loss": 5.3407,
      "mean_token_accuracy": 0.18301501870155334,
      "num_tokens": 61627420.0,
      "step": 29700
    },
    {
      "entropy": 5.6288961410522464,
      "epoch": 2.694575471698113,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004275822002727824,
      "loss": 5.296,
      "mean_token_accuracy": 0.18012533336877823,
      "num_tokens": 61638315.0,
      "step": 29705
    },
    {
      "entropy": 5.633039426803589,
      "epoch": 2.6950290275761972,
      "grad_norm": 1.9375,
      "learning_rate": 0.00042755841899562544,
      "loss": 5.2735,
      "mean_token_accuracy": 0.18568748980760574,
      "num_tokens": 61649294.0,
      "step": 29710
    },
    {
      "entropy": 5.7073616027832035,
      "epoch": 2.6954825834542815,
      "grad_norm": 1.5625,
      "learning_rate": 0.00042753463456352155,
      "loss": 5.323,
      "mean_token_accuracy": 0.1784167155623436,
      "num_tokens": 61660312.0,
      "step": 29715
    },
    {
      "entropy": 5.603904151916504,
      "epoch": 2.6959361393323658,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004275108469769625,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1869973972439766,
      "num_tokens": 61670539.0,
      "step": 29720
    },
    {
      "entropy": 5.61866717338562,
      "epoch": 2.69638969521045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042748705623644025,
      "loss": 5.216,
      "mean_token_accuracy": 0.1898484408855438,
      "num_tokens": 61679613.0,
      "step": 29725
    },
    {
      "entropy": 5.655160522460937,
      "epoch": 2.6968432510885343,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004274632623424468,
      "loss": 5.2953,
      "mean_token_accuracy": 0.1821514517068863,
      "num_tokens": 61689714.0,
      "step": 29730
    },
    {
      "entropy": 5.692232608795166,
      "epoch": 2.6972968069666186,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00042743946529547435,
      "loss": 5.3691,
      "mean_token_accuracy": 0.17710330933332444,
      "num_tokens": 61700701.0,
      "step": 29735
    },
    {
      "entropy": 5.677762508392334,
      "epoch": 2.6977503628447024,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004274156650960148,
      "loss": 5.3293,
      "mean_token_accuracy": 0.18232771009206772,
      "num_tokens": 61711403.0,
      "step": 29740
    },
    {
      "entropy": 5.539693689346313,
      "epoch": 2.6982039187227866,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004273918617445605,
      "loss": 5.1216,
      "mean_token_accuracy": 0.19457923322916032,
      "num_tokens": 61721787.0,
      "step": 29745
    },
    {
      "entropy": 5.656213855743408,
      "epoch": 2.698657474600871,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042736805524160363,
      "loss": 5.3863,
      "mean_token_accuracy": 0.17917137444019318,
      "num_tokens": 61732702.0,
      "step": 29750
    },
    {
      "entropy": 5.638692855834961,
      "epoch": 2.6991110304789547,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042734424558763656,
      "loss": 5.29,
      "mean_token_accuracy": 0.18698461204767228,
      "num_tokens": 61743477.0,
      "step": 29755
    },
    {
      "entropy": 5.66146297454834,
      "epoch": 2.699564586357039,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004273204327831516,
      "loss": 5.346,
      "mean_token_accuracy": 0.17840707153081894,
      "num_tokens": 61753620.0,
      "step": 29760
    },
    {
      "entropy": 5.657201862335205,
      "epoch": 2.7000181422351233,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004272966168286414,
      "loss": 5.2609,
      "mean_token_accuracy": 0.18725234270095825,
      "num_tokens": 61763678.0,
      "step": 29765
    },
    {
      "entropy": 5.679097127914429,
      "epoch": 2.7004716981132075,
      "grad_norm": 2.1875,
      "learning_rate": 0.00042727279772459837,
      "loss": 5.366,
      "mean_token_accuracy": 0.17878746390342712,
      "num_tokens": 61774419.0,
      "step": 29770
    },
    {
      "entropy": 5.714169597625732,
      "epoch": 2.700925253991292,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000427248975471515,
      "loss": 5.3409,
      "mean_token_accuracy": 0.18085113763809205,
      "num_tokens": 61785546.0,
      "step": 29775
    },
    {
      "entropy": 5.657928037643432,
      "epoch": 2.701378809869376,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042722515006988397,
      "loss": 5.2586,
      "mean_token_accuracy": 0.1843740835785866,
      "num_tokens": 61795846.0,
      "step": 29780
    },
    {
      "entropy": 5.613763093948364,
      "epoch": 2.7018323657474603,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004272013215201982,
      "loss": 5.3065,
      "mean_token_accuracy": 0.18599528819322586,
      "num_tokens": 61806440.0,
      "step": 29785
    },
    {
      "entropy": 5.601963138580322,
      "epoch": 2.702285921625544,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042717748982295015,
      "loss": 5.2441,
      "mean_token_accuracy": 0.18593545258045197,
      "num_tokens": 61815371.0,
      "step": 29790
    },
    {
      "entropy": 5.701010608673096,
      "epoch": 2.7027394775036284,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004271536549786329,
      "loss": 5.2299,
      "mean_token_accuracy": 0.18611503541469573,
      "num_tokens": 61825225.0,
      "step": 29795
    },
    {
      "entropy": 5.604143381118774,
      "epoch": 2.7031930333817127,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004271298169877393,
      "loss": 5.3256,
      "mean_token_accuracy": 0.18344642221927643,
      "num_tokens": 61836705.0,
      "step": 29800
    },
    {
      "entropy": 5.584060573577881,
      "epoch": 2.703646589259797,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004271059758507624,
      "loss": 5.2697,
      "mean_token_accuracy": 0.18637956082820892,
      "num_tokens": 61847082.0,
      "step": 29805
    },
    {
      "entropy": 5.635790586471558,
      "epoch": 2.7041001451378808,
      "grad_norm": 1.5,
      "learning_rate": 0.00042708213156819505,
      "loss": 5.2777,
      "mean_token_accuracy": 0.1847589522600174,
      "num_tokens": 61857130.0,
      "step": 29810
    },
    {
      "entropy": 5.709354257583618,
      "epoch": 2.704553701015965,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004270582841405305,
      "loss": 5.3679,
      "mean_token_accuracy": 0.17656695544719697,
      "num_tokens": 61866796.0,
      "step": 29815
    },
    {
      "entropy": 5.696286058425903,
      "epoch": 2.7050072568940493,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004270344335682618,
      "loss": 5.2975,
      "mean_token_accuracy": 0.17951156795024872,
      "num_tokens": 61876356.0,
      "step": 29820
    },
    {
      "entropy": 5.611270475387573,
      "epoch": 2.7054608127721336,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042701057985188235,
      "loss": 5.3191,
      "mean_token_accuracy": 0.17970592230558396,
      "num_tokens": 61887374.0,
      "step": 29825
    },
    {
      "entropy": 5.699968147277832,
      "epoch": 2.705914368650218,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004269867229918854,
      "loss": 5.352,
      "mean_token_accuracy": 0.18317507952451706,
      "num_tokens": 61897376.0,
      "step": 29830
    },
    {
      "entropy": 5.711379528045654,
      "epoch": 2.706367924528302,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042696286298876424,
      "loss": 5.3842,
      "mean_token_accuracy": 0.17471120357513428,
      "num_tokens": 61908212.0,
      "step": 29835
    },
    {
      "entropy": 5.7073322296142575,
      "epoch": 2.706821480406386,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004269389998430124,
      "loss": 5.3912,
      "mean_token_accuracy": 0.17121116071939468,
      "num_tokens": 61920699.0,
      "step": 29840
    },
    {
      "entropy": 5.667870092391968,
      "epoch": 2.70727503628447,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004269151335551233,
      "loss": 5.3232,
      "mean_token_accuracy": 0.18353310227394104,
      "num_tokens": 61931556.0,
      "step": 29845
    },
    {
      "entropy": 5.723425436019897,
      "epoch": 2.7077285921625545,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004268912641255905,
      "loss": 5.4249,
      "mean_token_accuracy": 0.183373361825943,
      "num_tokens": 61941483.0,
      "step": 29850
    },
    {
      "entropy": 5.654325532913208,
      "epoch": 2.7081821480406387,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042686739155490763,
      "loss": 5.2718,
      "mean_token_accuracy": 0.18705326020717622,
      "num_tokens": 61951262.0,
      "step": 29855
    },
    {
      "entropy": 5.706742143630981,
      "epoch": 2.7086357039187225,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004268435158435684,
      "loss": 5.3572,
      "mean_token_accuracy": 0.18166809231042863,
      "num_tokens": 61962423.0,
      "step": 29860
    },
    {
      "entropy": 5.718538665771485,
      "epoch": 2.709089259796807,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042681963699206667,
      "loss": 5.421,
      "mean_token_accuracy": 0.17324435859918594,
      "num_tokens": 61973276.0,
      "step": 29865
    },
    {
      "entropy": 5.603187370300293,
      "epoch": 2.709542815674891,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042679575500089605,
      "loss": 5.2265,
      "mean_token_accuracy": 0.18578286319971085,
      "num_tokens": 61984369.0,
      "step": 29870
    },
    {
      "entropy": 5.5789501667022705,
      "epoch": 2.7099963715529753,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004267718698705506,
      "loss": 5.2209,
      "mean_token_accuracy": 0.18316350877285004,
      "num_tokens": 61994224.0,
      "step": 29875
    },
    {
      "entropy": 5.6232977390289305,
      "epoch": 2.7104499274310596,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004267479816015242,
      "loss": 5.2961,
      "mean_token_accuracy": 0.19032059907913207,
      "num_tokens": 62004546.0,
      "step": 29880
    },
    {
      "entropy": 5.6909956455230715,
      "epoch": 2.710903483309144,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004267240901943109,
      "loss": 5.3015,
      "mean_token_accuracy": 0.18977024853229524,
      "num_tokens": 62013924.0,
      "step": 29885
    },
    {
      "entropy": 5.653233575820923,
      "epoch": 2.7113570391872277,
      "grad_norm": 2.3125,
      "learning_rate": 0.00042670019564940474,
      "loss": 5.2942,
      "mean_token_accuracy": 0.17719453126192092,
      "num_tokens": 62026150.0,
      "step": 29890
    },
    {
      "entropy": 5.673180437088012,
      "epoch": 2.711810595065312,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004266762979672998,
      "loss": 5.2933,
      "mean_token_accuracy": 0.1829736739397049,
      "num_tokens": 62036203.0,
      "step": 29895
    },
    {
      "entropy": 5.60765905380249,
      "epoch": 2.7122641509433962,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042665239714849044,
      "loss": 5.2917,
      "mean_token_accuracy": 0.18055867701768874,
      "num_tokens": 62046961.0,
      "step": 29900
    },
    {
      "entropy": 5.603221035003662,
      "epoch": 2.7127177068214805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042662849319347073,
      "loss": 5.2518,
      "mean_token_accuracy": 0.1833755850791931,
      "num_tokens": 62057365.0,
      "step": 29905
    },
    {
      "entropy": 5.653638553619385,
      "epoch": 2.7131712626995643,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042660458610273527,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1852399006485939,
      "num_tokens": 62067934.0,
      "step": 29910
    },
    {
      "entropy": 5.659613227844238,
      "epoch": 2.7136248185776486,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004265806758767783,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17986100614070893,
      "num_tokens": 62077203.0,
      "step": 29915
    },
    {
      "entropy": 5.653164291381836,
      "epoch": 2.714078374455733,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00042655676251609435,
      "loss": 5.2774,
      "mean_token_accuracy": 0.1849885791540146,
      "num_tokens": 62087262.0,
      "step": 29920
    },
    {
      "entropy": 5.5955376625061035,
      "epoch": 2.714531930333817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042653284602117795,
      "loss": 5.3141,
      "mean_token_accuracy": 0.18430437594652177,
      "num_tokens": 62097931.0,
      "step": 29925
    },
    {
      "entropy": 5.670137023925781,
      "epoch": 2.7149854862119014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004265089263925236,
      "loss": 5.3671,
      "mean_token_accuracy": 0.178787399828434,
      "num_tokens": 62109478.0,
      "step": 29930
    },
    {
      "entropy": 5.754874563217163,
      "epoch": 2.7154390420899857,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000426485003630626,
      "loss": 5.3595,
      "mean_token_accuracy": 0.17952117770910264,
      "num_tokens": 62120735.0,
      "step": 29935
    },
    {
      "entropy": 5.657505655288697,
      "epoch": 2.71589259796807,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004264610777359801,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18806450068950653,
      "num_tokens": 62130950.0,
      "step": 29940
    },
    {
      "entropy": 5.637121248245239,
      "epoch": 2.7163461538461537,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004264371487090804,
      "loss": 5.3012,
      "mean_token_accuracy": 0.1864370658993721,
      "num_tokens": 62141351.0,
      "step": 29945
    },
    {
      "entropy": 5.526107454299927,
      "epoch": 2.716799709724238,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042641321655042185,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18333253562450408,
      "num_tokens": 62151388.0,
      "step": 29950
    },
    {
      "entropy": 5.684361457824707,
      "epoch": 2.7172532656023223,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004263892812604995,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1799633100628853,
      "num_tokens": 62162809.0,
      "step": 29955
    },
    {
      "entropy": 5.691260766983032,
      "epoch": 2.7177068214804065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042636534283980814,
      "loss": 5.2908,
      "mean_token_accuracy": 0.18591542094945906,
      "num_tokens": 62172738.0,
      "step": 29960
    },
    {
      "entropy": 5.699423217773438,
      "epoch": 2.7181603773584904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000426341401288843,
      "loss": 5.4326,
      "mean_token_accuracy": 0.17712092250585557,
      "num_tokens": 62182508.0,
      "step": 29965
    },
    {
      "entropy": 5.690962553024292,
      "epoch": 2.7186139332365746,
      "grad_norm": 2.3125,
      "learning_rate": 0.00042631745660809905,
      "loss": 5.3801,
      "mean_token_accuracy": 0.1724992111325264,
      "num_tokens": 62192873.0,
      "step": 29970
    },
    {
      "entropy": 5.701973819732666,
      "epoch": 2.719067489114659,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042629350879807155,
      "loss": 5.4222,
      "mean_token_accuracy": 0.17426529079675673,
      "num_tokens": 62204727.0,
      "step": 29975
    },
    {
      "entropy": 5.674516439437866,
      "epoch": 2.719521044992743,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042626955785925576,
      "loss": 5.2785,
      "mean_token_accuracy": 0.18281429558992385,
      "num_tokens": 62214268.0,
      "step": 29980
    },
    {
      "entropy": 5.67451663017273,
      "epoch": 2.7199746008708274,
      "grad_norm": 2.75,
      "learning_rate": 0.00042624560379214693,
      "loss": 5.3361,
      "mean_token_accuracy": 0.18441876471042634,
      "num_tokens": 62224756.0,
      "step": 29985
    },
    {
      "entropy": 5.6508598804473875,
      "epoch": 2.7204281567489117,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004262216465972405,
      "loss": 5.2798,
      "mean_token_accuracy": 0.18600487262010573,
      "num_tokens": 62234065.0,
      "step": 29990
    },
    {
      "entropy": 5.681738424301147,
      "epoch": 2.7208817126269955,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004261976862750318,
      "loss": 5.3062,
      "mean_token_accuracy": 0.18084548711776732,
      "num_tokens": 62243732.0,
      "step": 29995
    },
    {
      "entropy": 5.6909508228302,
      "epoch": 2.72133526850508,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004261737228260165,
      "loss": 5.3288,
      "mean_token_accuracy": 0.18177279233932495,
      "num_tokens": 62253461.0,
      "step": 30000
    },
    {
      "epoch": 2.72133526850508,
      "eval_entropy": 5.4138571566586045,
      "eval_loss": 5.3475494384765625,
      "eval_mean_token_accuracy": 0.18908811393889233,
      "eval_num_tokens": 62253461.0,
      "eval_runtime": 26.7657,
      "eval_samples_per_second": 1321.094,
      "eval_steps_per_second": 165.137,
      "step": 30000
    },
    {
      "entropy": 5.676731729507447,
      "epoch": 2.721788824383164,
      "grad_norm": 1.375,
      "learning_rate": 0.0004261497562506901,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1811368003487587,
      "num_tokens": 62263830.0,
      "step": 30005
    },
    {
      "entropy": 5.567401266098022,
      "epoch": 2.7222423802612483,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004261257865495483,
      "loss": 5.2387,
      "mean_token_accuracy": 0.18993799835443498,
      "num_tokens": 62274383.0,
      "step": 30010
    },
    {
      "entropy": 5.714111375808716,
      "epoch": 2.722695936139332,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004261018137230865,
      "loss": 5.4333,
      "mean_token_accuracy": 0.16888678818941116,
      "num_tokens": 62286476.0,
      "step": 30015
    },
    {
      "entropy": 5.7469319820404055,
      "epoch": 2.7231494920174164,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00042607783777180085,
      "loss": 5.3286,
      "mean_token_accuracy": 0.18330759108066558,
      "num_tokens": 62296192.0,
      "step": 30020
    },
    {
      "entropy": 5.56952133178711,
      "epoch": 2.7236030478955007,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042605385869618693,
      "loss": 5.2399,
      "mean_token_accuracy": 0.19272126257419586,
      "num_tokens": 62305300.0,
      "step": 30025
    },
    {
      "entropy": 5.658433818817139,
      "epoch": 2.724056603773585,
      "grad_norm": 1.5,
      "learning_rate": 0.0004260298764967408,
      "loss": 5.298,
      "mean_token_accuracy": 0.18406918197870253,
      "num_tokens": 62316802.0,
      "step": 30030
    },
    {
      "entropy": 5.589628982543945,
      "epoch": 2.724510159651669,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004260058911739583,
      "loss": 5.262,
      "mean_token_accuracy": 0.18361200094223024,
      "num_tokens": 62326444.0,
      "step": 30035
    },
    {
      "entropy": 5.6635932445526125,
      "epoch": 2.7249637155297535,
      "grad_norm": 2.078125,
      "learning_rate": 0.00042598190272833547,
      "loss": 5.237,
      "mean_token_accuracy": 0.18559882491827012,
      "num_tokens": 62336872.0,
      "step": 30040
    },
    {
      "entropy": 5.528777265548706,
      "epoch": 2.7254172714078373,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004259579111603685,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19141251891851424,
      "num_tokens": 62346771.0,
      "step": 30045
    },
    {
      "entropy": 5.633358573913574,
      "epoch": 2.7258708272859216,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042593391647055334,
      "loss": 5.3111,
      "mean_token_accuracy": 0.18723763078451156,
      "num_tokens": 62357503.0,
      "step": 30050
    },
    {
      "entropy": 5.583203363418579,
      "epoch": 2.726324383164006,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004259099186593864,
      "loss": 5.181,
      "mean_token_accuracy": 0.1916812151670456,
      "num_tokens": 62367664.0,
      "step": 30055
    },
    {
      "entropy": 5.6200672626495365,
      "epoch": 2.72677793904209,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004258859177273639,
      "loss": 5.2503,
      "mean_token_accuracy": 0.18175522089004517,
      "num_tokens": 62376924.0,
      "step": 30060
    },
    {
      "entropy": 5.527755784988403,
      "epoch": 2.727231494920174,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004258619136749821,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1901668682694435,
      "num_tokens": 62387529.0,
      "step": 30065
    },
    {
      "entropy": 5.605148077011108,
      "epoch": 2.727685050798258,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004258379065027375,
      "loss": 5.2574,
      "mean_token_accuracy": 0.1842309206724167,
      "num_tokens": 62397833.0,
      "step": 30070
    },
    {
      "entropy": 5.733967924118042,
      "epoch": 2.7281386066763424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004258138962111266,
      "loss": 5.3311,
      "mean_token_accuracy": 0.18461036086082458,
      "num_tokens": 62409022.0,
      "step": 30075
    },
    {
      "entropy": 5.6208549499511715,
      "epoch": 2.7285921625544267,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004257898828006459,
      "loss": 5.2896,
      "mean_token_accuracy": 0.17831607460975646,
      "num_tokens": 62419105.0,
      "step": 30080
    },
    {
      "entropy": 5.6109882354736325,
      "epoch": 2.729045718432511,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004257658662717919,
      "loss": 5.2809,
      "mean_token_accuracy": 0.18693211674690247,
      "num_tokens": 62428371.0,
      "step": 30085
    },
    {
      "entropy": 5.681070137023926,
      "epoch": 2.7294992743105952,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004257418466250615,
      "loss": 5.254,
      "mean_token_accuracy": 0.18481141179800034,
      "num_tokens": 62439301.0,
      "step": 30090
    },
    {
      "entropy": 5.636351108551025,
      "epoch": 2.7299528301886795,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042571782386095124,
      "loss": 5.2625,
      "mean_token_accuracy": 0.18399045914411544,
      "num_tokens": 62449450.0,
      "step": 30095
    },
    {
      "entropy": 5.644216871261596,
      "epoch": 2.7304063860667633,
      "grad_norm": 1.90625,
      "learning_rate": 0.00042569379797995793,
      "loss": 5.2716,
      "mean_token_accuracy": 0.17808563858270646,
      "num_tokens": 62461051.0,
      "step": 30100
    },
    {
      "entropy": 5.608516216278076,
      "epoch": 2.7308599419448476,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004256697689825785,
      "loss": 5.216,
      "mean_token_accuracy": 0.18893166184425353,
      "num_tokens": 62471755.0,
      "step": 30105
    },
    {
      "entropy": 5.5954859256744385,
      "epoch": 2.731313497822932,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042564573686930987,
      "loss": 5.245,
      "mean_token_accuracy": 0.18745509088039397,
      "num_tokens": 62482060.0,
      "step": 30110
    },
    {
      "entropy": 5.606687879562378,
      "epoch": 2.7317670537010157,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000425621701640649,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18465204983949662,
      "num_tokens": 62493735.0,
      "step": 30115
    },
    {
      "entropy": 5.61294846534729,
      "epoch": 2.7322206095791,
      "grad_norm": 1.765625,
      "learning_rate": 0.000425597663297093,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17719956040382384,
      "num_tokens": 62504582.0,
      "step": 30120
    },
    {
      "entropy": 5.5657721042633055,
      "epoch": 2.7326741654571842,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004255736218391389,
      "loss": 5.2512,
      "mean_token_accuracy": 0.1819467216730118,
      "num_tokens": 62516665.0,
      "step": 30125
    },
    {
      "entropy": 5.643116760253906,
      "epoch": 2.7331277213352685,
      "grad_norm": 1.6875,
      "learning_rate": 0.00042554957726728394,
      "loss": 5.3021,
      "mean_token_accuracy": 0.18422954082489013,
      "num_tokens": 62527745.0,
      "step": 30130
    },
    {
      "entropy": 5.6303791999816895,
      "epoch": 2.7335812772133528,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004255255295820253,
      "loss": 5.2659,
      "mean_token_accuracy": 0.1894255056977272,
      "num_tokens": 62538622.0,
      "step": 30135
    },
    {
      "entropy": 5.5936455726623535,
      "epoch": 2.734034833091437,
      "grad_norm": 4.90625,
      "learning_rate": 0.00042550147878386044,
      "loss": 5.2648,
      "mean_token_accuracy": 0.18627216815948486,
      "num_tokens": 62548302.0,
      "step": 30140
    },
    {
      "entropy": 5.564675188064575,
      "epoch": 2.7344883889695213,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004254774248732867,
      "loss": 5.196,
      "mean_token_accuracy": 0.19341716468334197,
      "num_tokens": 62557150.0,
      "step": 30145
    },
    {
      "entropy": 5.763157606124878,
      "epoch": 2.734941944847605,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00042545336785080126,
      "loss": 5.4679,
      "mean_token_accuracy": 0.17049481794238092,
      "num_tokens": 62568291.0,
      "step": 30150
    },
    {
      "entropy": 5.645722675323486,
      "epoch": 2.7353955007256894,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000425429307716902,
      "loss": 5.2817,
      "mean_token_accuracy": 0.18199051469564437,
      "num_tokens": 62579351.0,
      "step": 30155
    },
    {
      "entropy": 5.5300665378570555,
      "epoch": 2.7358490566037736,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042540524447208625,
      "loss": 5.1372,
      "mean_token_accuracy": 0.19174949824810028,
      "num_tokens": 62588614.0,
      "step": 30160
    },
    {
      "entropy": 5.65567626953125,
      "epoch": 2.736302612481858,
      "grad_norm": 1.609375,
      "learning_rate": 0.00042538117811685174,
      "loss": 5.3046,
      "mean_token_accuracy": 0.18835723102092744,
      "num_tokens": 62598844.0,
      "step": 30165
    },
    {
      "entropy": 5.633611011505127,
      "epoch": 2.7367561683599417,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042535710865169616,
      "loss": 5.3168,
      "mean_token_accuracy": 0.1795751243829727,
      "num_tokens": 62608722.0,
      "step": 30170
    },
    {
      "entropy": 5.588191270828247,
      "epoch": 2.737209724238026,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004253330360771173,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18328244537115096,
      "num_tokens": 62618712.0,
      "step": 30175
    },
    {
      "entropy": 5.6867413997650145,
      "epoch": 2.7376632801161103,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042530896039361284,
      "loss": 5.3532,
      "mean_token_accuracy": 0.17660048902034758,
      "num_tokens": 62629329.0,
      "step": 30180
    },
    {
      "entropy": 5.669504976272583,
      "epoch": 2.7381168359941945,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042528488160168077,
      "loss": 5.3305,
      "mean_token_accuracy": 0.17762513160705568,
      "num_tokens": 62639644.0,
      "step": 30185
    },
    {
      "entropy": 5.644729804992676,
      "epoch": 2.738570391872279,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042526079970181915,
      "loss": 5.3454,
      "mean_token_accuracy": 0.1756424069404602,
      "num_tokens": 62650979.0,
      "step": 30190
    },
    {
      "entropy": 5.680493259429932,
      "epoch": 2.739023947750363,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004252367146945259,
      "loss": 5.374,
      "mean_token_accuracy": 0.17984914183616638,
      "num_tokens": 62661163.0,
      "step": 30195
    },
    {
      "entropy": 5.671449136734009,
      "epoch": 2.739477503628447,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042521262658029905,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17870650291442872,
      "num_tokens": 62672371.0,
      "step": 30200
    },
    {
      "entropy": 5.629363775253296,
      "epoch": 2.739931059506531,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042518853535963673,
      "loss": 5.2285,
      "mean_token_accuracy": 0.1868928611278534,
      "num_tokens": 62682521.0,
      "step": 30205
    },
    {
      "entropy": 5.60608286857605,
      "epoch": 2.7403846153846154,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00042516444103303736,
      "loss": 5.1674,
      "mean_token_accuracy": 0.1891463503241539,
      "num_tokens": 62692420.0,
      "step": 30210
    },
    {
      "entropy": 5.616644859313965,
      "epoch": 2.7408381712626997,
      "grad_norm": 2.984375,
      "learning_rate": 0.00042514034360099907,
      "loss": 5.2501,
      "mean_token_accuracy": 0.18586984127759934,
      "num_tokens": 62702686.0,
      "step": 30215
    },
    {
      "entropy": 5.595649671554566,
      "epoch": 2.7412917271407835,
      "grad_norm": 1.859375,
      "learning_rate": 0.00042511624306402015,
      "loss": 5.2686,
      "mean_token_accuracy": 0.18478535562753678,
      "num_tokens": 62713262.0,
      "step": 30220
    },
    {
      "entropy": 5.654236650466919,
      "epoch": 2.7417452830188678,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004250921394225992,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18996575623750686,
      "num_tokens": 62724460.0,
      "step": 30225
    },
    {
      "entropy": 5.636218023300171,
      "epoch": 2.742198838896952,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004250680326772343,
      "loss": 5.2973,
      "mean_token_accuracy": 0.18785660713911057,
      "num_tokens": 62734190.0,
      "step": 30230
    },
    {
      "entropy": 5.640984392166137,
      "epoch": 2.7426523947750363,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004250439228284244,
      "loss": 5.2506,
      "mean_token_accuracy": 0.18169442862272261,
      "num_tokens": 62744323.0,
      "step": 30235
    },
    {
      "entropy": 5.635057735443115,
      "epoch": 2.7431059506531206,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042501980987666807,
      "loss": 5.2489,
      "mean_token_accuracy": 0.1853928357362747,
      "num_tokens": 62753868.0,
      "step": 30240
    },
    {
      "entropy": 5.656902217864991,
      "epoch": 2.743559506531205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004249956938224636,
      "loss": 5.3481,
      "mean_token_accuracy": 0.18561861366033555,
      "num_tokens": 62763954.0,
      "step": 30245
    },
    {
      "entropy": 5.612594509124756,
      "epoch": 2.7440130624092887,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042497157466631015,
      "loss": 5.2797,
      "mean_token_accuracy": 0.18680256158113479,
      "num_tokens": 62774116.0,
      "step": 30250
    },
    {
      "entropy": 5.7304847717285154,
      "epoch": 2.744466618287373,
      "grad_norm": 2.234375,
      "learning_rate": 0.00042494745240870626,
      "loss": 5.3111,
      "mean_token_accuracy": 0.1793008714914322,
      "num_tokens": 62784939.0,
      "step": 30255
    },
    {
      "entropy": 5.662718439102173,
      "epoch": 2.744920174165457,
      "grad_norm": 1.5625,
      "learning_rate": 0.00042492332705015075,
      "loss": 5.2933,
      "mean_token_accuracy": 0.18353525549173355,
      "num_tokens": 62794988.0,
      "step": 30260
    },
    {
      "entropy": 5.612292528152466,
      "epoch": 2.7453737300435415,
      "grad_norm": 2.03125,
      "learning_rate": 0.00042489919859114273,
      "loss": 5.2955,
      "mean_token_accuracy": 0.19177430719137192,
      "num_tokens": 62804984.0,
      "step": 30265
    },
    {
      "entropy": 5.6667743682861325,
      "epoch": 2.7458272859216253,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042487506703218104,
      "loss": 5.2778,
      "mean_token_accuracy": 0.175430092215538,
      "num_tokens": 62816401.0,
      "step": 30270
    },
    {
      "entropy": 5.61528844833374,
      "epoch": 2.7462808417997095,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004248509323737648,
      "loss": 5.2353,
      "mean_token_accuracy": 0.18313169181346894,
      "num_tokens": 62826411.0,
      "step": 30275
    },
    {
      "entropy": 5.657751512527466,
      "epoch": 2.746734397677794,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004248267946163931,
      "loss": 5.3289,
      "mean_token_accuracy": 0.1828339859843254,
      "num_tokens": 62837228.0,
      "step": 30280
    },
    {
      "entropy": 5.573294734954834,
      "epoch": 2.747187953555878,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00042480265376056506,
      "loss": 5.1899,
      "mean_token_accuracy": 0.19190959632396698,
      "num_tokens": 62847852.0,
      "step": 30285
    },
    {
      "entropy": 5.598988199234009,
      "epoch": 2.7476415094339623,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042477850980678,
      "loss": 5.3125,
      "mean_token_accuracy": 0.18047351837158204,
      "num_tokens": 62859516.0,
      "step": 30290
    },
    {
      "entropy": 5.68572416305542,
      "epoch": 2.7480950653120466,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004247543627555372,
      "loss": 5.2724,
      "mean_token_accuracy": 0.18467389792203903,
      "num_tokens": 62869757.0,
      "step": 30295
    },
    {
      "entropy": 5.568242406845092,
      "epoch": 2.748548621190131,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000424730212607336,
      "loss": 5.3547,
      "mean_token_accuracy": 0.18061317950487138,
      "num_tokens": 62879618.0,
      "step": 30300
    },
    {
      "entropy": 5.622831964492798,
      "epoch": 2.7490021770682147,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004247060593626758,
      "loss": 5.3037,
      "mean_token_accuracy": 0.17836359590291978,
      "num_tokens": 62890491.0,
      "step": 30305
    },
    {
      "entropy": 5.7253069400787355,
      "epoch": 2.749455732946299,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004246819030220562,
      "loss": 5.3082,
      "mean_token_accuracy": 0.19053157567977905,
      "num_tokens": 62899954.0,
      "step": 30310
    },
    {
      "entropy": 5.617915201187134,
      "epoch": 2.7499092888243832,
      "grad_norm": 1.5,
      "learning_rate": 0.0004246577435859767,
      "loss": 5.3498,
      "mean_token_accuracy": 0.18538357615470885,
      "num_tokens": 62911857.0,
      "step": 30315
    },
    {
      "entropy": 5.640580511093139,
      "epoch": 2.7503628447024675,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004246335810549369,
      "loss": 5.2627,
      "mean_token_accuracy": 0.1877265453338623,
      "num_tokens": 62922023.0,
      "step": 30320
    },
    {
      "entropy": 5.6744469165802,
      "epoch": 2.7508164005805513,
      "grad_norm": 1.984375,
      "learning_rate": 0.00042460941542943656,
      "loss": 5.3213,
      "mean_token_accuracy": 0.1908768430352211,
      "num_tokens": 62932845.0,
      "step": 30325
    },
    {
      "entropy": 5.669528961181641,
      "epoch": 2.7512699564586356,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004245852467099753,
      "loss": 5.2677,
      "mean_token_accuracy": 0.1860465496778488,
      "num_tokens": 62942881.0,
      "step": 30330
    },
    {
      "entropy": 5.614163970947265,
      "epoch": 2.75172351233672,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004245610748970531,
      "loss": 5.2561,
      "mean_token_accuracy": 0.18284743130207062,
      "num_tokens": 62953957.0,
      "step": 30335
    },
    {
      "entropy": 5.5648314476013185,
      "epoch": 2.752177068214804,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042453689999116974,
      "loss": 5.2496,
      "mean_token_accuracy": 0.19186995476484298,
      "num_tokens": 62964803.0,
      "step": 30340
    },
    {
      "entropy": 5.585123109817505,
      "epoch": 2.7526306240928884,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004245127219928252,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1873278558254242,
      "num_tokens": 62975477.0,
      "step": 30345
    },
    {
      "entropy": 5.606790828704834,
      "epoch": 2.7530841799709727,
      "grad_norm": 1.9375,
      "learning_rate": 0.00042448854090251944,
      "loss": 5.2803,
      "mean_token_accuracy": 0.1866640791296959,
      "num_tokens": 62985430.0,
      "step": 30350
    },
    {
      "entropy": 5.624991321563721,
      "epoch": 2.7535377358490565,
      "grad_norm": 1.609375,
      "learning_rate": 0.00042446435672075254,
      "loss": 5.3061,
      "mean_token_accuracy": 0.18382263481616973,
      "num_tokens": 62997493.0,
      "step": 30355
    },
    {
      "entropy": 5.6899213790893555,
      "epoch": 2.7539912917271407,
      "grad_norm": 1.5,
      "learning_rate": 0.00042444016944802467,
      "loss": 5.4063,
      "mean_token_accuracy": 0.18174235969781877,
      "num_tokens": 63007573.0,
      "step": 30360
    },
    {
      "entropy": 5.678098249435425,
      "epoch": 2.754444847605225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042441597908483603,
      "loss": 5.3389,
      "mean_token_accuracy": 0.18652136176824569,
      "num_tokens": 63018372.0,
      "step": 30365
    },
    {
      "entropy": 5.606819772720337,
      "epoch": 2.7548984034833093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004243917856316869,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18665675669908524,
      "num_tokens": 63028632.0,
      "step": 30370
    },
    {
      "entropy": 5.631274032592773,
      "epoch": 2.755351959361393,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042436758908907747,
      "loss": 5.248,
      "mean_token_accuracy": 0.184917214512825,
      "num_tokens": 63038510.0,
      "step": 30375
    },
    {
      "entropy": 5.753591585159302,
      "epoch": 2.7558055152394774,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042434338945750833,
      "loss": 5.4409,
      "mean_token_accuracy": 0.16660493910312651,
      "num_tokens": 63049375.0,
      "step": 30380
    },
    {
      "entropy": 5.6561953067779545,
      "epoch": 2.7562590711175616,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042431918673747974,
      "loss": 5.2714,
      "mean_token_accuracy": 0.18413971066474916,
      "num_tokens": 63059422.0,
      "step": 30385
    },
    {
      "entropy": 5.696762466430664,
      "epoch": 2.756712626995646,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042429498092949245,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17345014810562134,
      "num_tokens": 63071101.0,
      "step": 30390
    },
    {
      "entropy": 5.584912872314453,
      "epoch": 2.75716618287373,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004242707720340468,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19049662947654725,
      "num_tokens": 63081104.0,
      "step": 30395
    },
    {
      "entropy": 5.634474754333496,
      "epoch": 2.7576197387518144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004242465600516436,
      "loss": 5.3038,
      "mean_token_accuracy": 0.1815878137946129,
      "num_tokens": 63091123.0,
      "step": 30400
    },
    {
      "entropy": 5.618531942367554,
      "epoch": 2.7580732946298983,
      "grad_norm": 2.0,
      "learning_rate": 0.00042422234498278353,
      "loss": 5.2753,
      "mean_token_accuracy": 0.1849986732006073,
      "num_tokens": 63101541.0,
      "step": 30405
    },
    {
      "entropy": 5.624760389328003,
      "epoch": 2.7585268505079825,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042419812682796727,
      "loss": 5.3495,
      "mean_token_accuracy": 0.1783716768026352,
      "num_tokens": 63112012.0,
      "step": 30410
    },
    {
      "entropy": 5.651214647293091,
      "epoch": 2.758980406386067,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004241739055876957,
      "loss": 5.2738,
      "mean_token_accuracy": 0.18606246560811995,
      "num_tokens": 63122021.0,
      "step": 30415
    },
    {
      "entropy": 5.730993795394897,
      "epoch": 2.759433962264151,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042414968126246985,
      "loss": 5.3853,
      "mean_token_accuracy": 0.1809280440211296,
      "num_tokens": 63132286.0,
      "step": 30420
    },
    {
      "entropy": 5.6197303295135494,
      "epoch": 2.759887518142235,
      "grad_norm": 1.6875,
      "learning_rate": 0.00042412545385279055,
      "loss": 5.2589,
      "mean_token_accuracy": 0.18917582482099532,
      "num_tokens": 63142159.0,
      "step": 30425
    },
    {
      "entropy": 5.599758195877075,
      "epoch": 2.760341074020319,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042410122335915883,
      "loss": 5.2309,
      "mean_token_accuracy": 0.18237110823392869,
      "num_tokens": 63152225.0,
      "step": 30430
    },
    {
      "entropy": 5.608938503265381,
      "epoch": 2.7607946298984034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004240769897820758,
      "loss": 5.265,
      "mean_token_accuracy": 0.18347158282995224,
      "num_tokens": 63162522.0,
      "step": 30435
    },
    {
      "entropy": 5.651372289657592,
      "epoch": 2.7612481857764877,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042405275312204266,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18953754305839537,
      "num_tokens": 63172094.0,
      "step": 30440
    },
    {
      "entropy": 5.623358154296875,
      "epoch": 2.761701741654572,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042402851337956066,
      "loss": 5.2852,
      "mean_token_accuracy": 0.18525506407022477,
      "num_tokens": 63182122.0,
      "step": 30445
    },
    {
      "entropy": 5.561495780944824,
      "epoch": 2.762155297532656,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042400427055513094,
      "loss": 5.2605,
      "mean_token_accuracy": 0.18234988003969194,
      "num_tokens": 63193174.0,
      "step": 30450
    },
    {
      "entropy": 5.662054061889648,
      "epoch": 2.7626088534107405,
      "grad_norm": 1.59375,
      "learning_rate": 0.000423980024649255,
      "loss": 5.2173,
      "mean_token_accuracy": 0.19433563351631164,
      "num_tokens": 63203698.0,
      "step": 30455
    },
    {
      "entropy": 5.52830023765564,
      "epoch": 2.7630624092888243,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004239557756624342,
      "loss": 5.1524,
      "mean_token_accuracy": 0.18901564925909042,
      "num_tokens": 63213343.0,
      "step": 30460
    },
    {
      "entropy": 5.66136531829834,
      "epoch": 2.7635159651669086,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042393152359516994,
      "loss": 5.3939,
      "mean_token_accuracy": 0.17771244496107103,
      "num_tokens": 63223962.0,
      "step": 30465
    },
    {
      "entropy": 5.627434730529785,
      "epoch": 2.763969521044993,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00042390726844796384,
      "loss": 5.2466,
      "mean_token_accuracy": 0.18378076702356339,
      "num_tokens": 63234279.0,
      "step": 30470
    },
    {
      "entropy": 5.629749727249146,
      "epoch": 2.7644230769230766,
      "grad_norm": 1.375,
      "learning_rate": 0.0004238830102213174,
      "loss": 5.3158,
      "mean_token_accuracy": 0.18360173106193542,
      "num_tokens": 63245567.0,
      "step": 30475
    },
    {
      "entropy": 5.609576892852783,
      "epoch": 2.764876632801161,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004238587489157325,
      "loss": 5.2429,
      "mean_token_accuracy": 0.19051066040992737,
      "num_tokens": 63256227.0,
      "step": 30480
    },
    {
      "entropy": 5.558459711074829,
      "epoch": 2.765330188679245,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004238344845317107,
      "loss": 5.1815,
      "mean_token_accuracy": 0.1920675292611122,
      "num_tokens": 63266618.0,
      "step": 30485
    },
    {
      "entropy": 5.567308330535889,
      "epoch": 2.7657837445573294,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004238102170697539,
      "loss": 5.2152,
      "mean_token_accuracy": 0.1919454574584961,
      "num_tokens": 63277130.0,
      "step": 30490
    },
    {
      "entropy": 5.528900003433227,
      "epoch": 2.7662373004354137,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004237859465303637,
      "loss": 5.1315,
      "mean_token_accuracy": 0.19415298849344254,
      "num_tokens": 63287351.0,
      "step": 30495
    },
    {
      "entropy": 5.5266670227050785,
      "epoch": 2.766690856313498,
      "grad_norm": 1.59375,
      "learning_rate": 0.00042376167291404234,
      "loss": 5.2393,
      "mean_token_accuracy": 0.18487464934587478,
      "num_tokens": 63298573.0,
      "step": 30500
    },
    {
      "entropy": 5.639365148544312,
      "epoch": 2.7671444121915822,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042373739622129166,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17985644787549973,
      "num_tokens": 63310740.0,
      "step": 30505
    },
    {
      "entropy": 5.6394671440124515,
      "epoch": 2.767597968069666,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042371311645261376,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18501786291599273,
      "num_tokens": 63321323.0,
      "step": 30510
    },
    {
      "entropy": 5.6367431640625,
      "epoch": 2.7680515239477503,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004236888336085107,
      "loss": 5.2777,
      "mean_token_accuracy": 0.1878033846616745,
      "num_tokens": 63332322.0,
      "step": 30515
    },
    {
      "entropy": 5.568124294281006,
      "epoch": 2.7685050798258346,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004236645476894846,
      "loss": 5.1563,
      "mean_token_accuracy": 0.1930684983730316,
      "num_tokens": 63342204.0,
      "step": 30520
    },
    {
      "entropy": 5.485630750656128,
      "epoch": 2.768958635703919,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004236402586960378,
      "loss": 5.2342,
      "mean_token_accuracy": 0.18924863934516906,
      "num_tokens": 63352460.0,
      "step": 30525
    },
    {
      "entropy": 5.544005298614502,
      "epoch": 2.7694121915820027,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004236159666286726,
      "loss": 5.2091,
      "mean_token_accuracy": 0.1873753100633621,
      "num_tokens": 63362474.0,
      "step": 30530
    },
    {
      "entropy": 5.650790214538574,
      "epoch": 2.769865747460087,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042359167148789135,
      "loss": 5.2623,
      "mean_token_accuracy": 0.19327131062746047,
      "num_tokens": 63372537.0,
      "step": 30535
    },
    {
      "entropy": 5.561816644668579,
      "epoch": 2.7703193033381712,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004235673732741964,
      "loss": 5.148,
      "mean_token_accuracy": 0.197308486700058,
      "num_tokens": 63382985.0,
      "step": 30540
    },
    {
      "entropy": 5.621789121627808,
      "epoch": 2.7707728592162555,
      "grad_norm": 2.28125,
      "learning_rate": 0.00042354307198809037,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18601517528295516,
      "num_tokens": 63393424.0,
      "step": 30545
    },
    {
      "entropy": 5.546093273162842,
      "epoch": 2.7712264150943398,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004235187676300757,
      "loss": 5.1288,
      "mean_token_accuracy": 0.19687415957450866,
      "num_tokens": 63403160.0,
      "step": 30550
    },
    {
      "entropy": 5.675464820861817,
      "epoch": 2.771679970972424,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004234944602006552,
      "loss": 5.2893,
      "mean_token_accuracy": 0.1865466132760048,
      "num_tokens": 63413001.0,
      "step": 30555
    },
    {
      "entropy": 5.603711032867432,
      "epoch": 2.772133526850508,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042347014970033124,
      "loss": 5.235,
      "mean_token_accuracy": 0.18701835721731186,
      "num_tokens": 63422735.0,
      "step": 30560
    },
    {
      "entropy": 5.483384370803833,
      "epoch": 2.772587082728592,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004234458361296069,
      "loss": 5.2288,
      "mean_token_accuracy": 0.18826694637537003,
      "num_tokens": 63432566.0,
      "step": 30565
    },
    {
      "entropy": 5.620724773406982,
      "epoch": 2.7730406386066764,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042342151948898476,
      "loss": 5.2462,
      "mean_token_accuracy": 0.18701807409524918,
      "num_tokens": 63442412.0,
      "step": 30570
    },
    {
      "entropy": 5.614898777008056,
      "epoch": 2.7734941944847606,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00042339719977896784,
      "loss": 5.2509,
      "mean_token_accuracy": 0.18985108733177186,
      "num_tokens": 63453018.0,
      "step": 30575
    },
    {
      "entropy": 5.629516077041626,
      "epoch": 2.7739477503628445,
      "grad_norm": 1.484375,
      "learning_rate": 0.000423372877000059,
      "loss": 5.2821,
      "mean_token_accuracy": 0.18105800747871398,
      "num_tokens": 63464286.0,
      "step": 30580
    },
    {
      "entropy": 5.575534868240356,
      "epoch": 2.7744013062409287,
      "grad_norm": 2.453125,
      "learning_rate": 0.00042334855115276117,
      "loss": 5.2695,
      "mean_token_accuracy": 0.18264537155628205,
      "num_tokens": 63475138.0,
      "step": 30585
    },
    {
      "entropy": 5.690752124786377,
      "epoch": 2.774854862119013,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00042332422223757756,
      "loss": 5.3258,
      "mean_token_accuracy": 0.18593557626008989,
      "num_tokens": 63486104.0,
      "step": 30590
    },
    {
      "entropy": 5.581053161621094,
      "epoch": 2.7753084179970973,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042329989025501123,
      "loss": 5.2435,
      "mean_token_accuracy": 0.18843038529157638,
      "num_tokens": 63496490.0,
      "step": 30595
    },
    {
      "entropy": 5.595835256576538,
      "epoch": 2.7757619738751815,
      "grad_norm": 2.375,
      "learning_rate": 0.00042327555520556536,
      "loss": 5.2071,
      "mean_token_accuracy": 0.1854623705148697,
      "num_tokens": 63506562.0,
      "step": 30600
    },
    {
      "entropy": 5.611049652099609,
      "epoch": 2.776215529753266,
      "grad_norm": 1.890625,
      "learning_rate": 0.00042325121708974324,
      "loss": 5.258,
      "mean_token_accuracy": 0.1913444444537163,
      "num_tokens": 63516641.0,
      "step": 30605
    },
    {
      "entropy": 5.633036470413208,
      "epoch": 2.7766690856313496,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004232268759080482,
      "loss": 5.241,
      "mean_token_accuracy": 0.19489616602659227,
      "num_tokens": 63526864.0,
      "step": 30610
    },
    {
      "entropy": 5.635004663467408,
      "epoch": 2.777122641509434,
      "grad_norm": 1.734375,
      "learning_rate": 0.00042320253166098354,
      "loss": 5.3288,
      "mean_token_accuracy": 0.18094661831855774,
      "num_tokens": 63537160.0,
      "step": 30615
    },
    {
      "entropy": 5.6362522602081295,
      "epoch": 2.777576197387518,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00042317818434905275,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18790933042764663,
      "num_tokens": 63547485.0,
      "step": 30620
    },
    {
      "entropy": 5.641354131698608,
      "epoch": 2.7780297532656024,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004231538339727594,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18496844172477722,
      "num_tokens": 63559369.0,
      "step": 30625
    },
    {
      "entropy": 5.746356010437012,
      "epoch": 2.7784833091436862,
      "grad_norm": 1.59375,
      "learning_rate": 0.00042312948053260704,
      "loss": 5.4149,
      "mean_token_accuracy": 0.17498500943183898,
      "num_tokens": 63570033.0,
      "step": 30630
    },
    {
      "entropy": 5.664302921295166,
      "epoch": 2.7789368650217705,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004231051240290992,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1873343989253044,
      "num_tokens": 63581274.0,
      "step": 30635
    },
    {
      "entropy": 5.5967864990234375,
      "epoch": 2.7793904208998548,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042308076446273963,
      "loss": 5.2541,
      "mean_token_accuracy": 0.1836978390812874,
      "num_tokens": 63590903.0,
      "step": 30640
    },
    {
      "entropy": 5.547576904296875,
      "epoch": 2.779843976777939,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00042305640183403214,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18250153660774232,
      "num_tokens": 63600982.0,
      "step": 30645
    },
    {
      "entropy": 5.572157716751098,
      "epoch": 2.7802975326560233,
      "grad_norm": 2.3125,
      "learning_rate": 0.00042303203614348057,
      "loss": 5.1602,
      "mean_token_accuracy": 0.189734947681427,
      "num_tokens": 63612411.0,
      "step": 30650
    },
    {
      "entropy": 5.5812419891357425,
      "epoch": 2.7807510885341076,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004230076673915888,
      "loss": 5.2241,
      "mean_token_accuracy": 0.19482938051223755,
      "num_tokens": 63622128.0,
      "step": 30655
    },
    {
      "entropy": 5.588881492614746,
      "epoch": 2.781204644412192,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042298329557886074,
      "loss": 5.2714,
      "mean_token_accuracy": 0.18950304687023162,
      "num_tokens": 63632047.0,
      "step": 30660
    },
    {
      "entropy": 5.7167600154876705,
      "epoch": 2.7816582002902757,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004229589207058003,
      "loss": 5.3168,
      "mean_token_accuracy": 0.1870155319571495,
      "num_tokens": 63642490.0,
      "step": 30665
    },
    {
      "entropy": 5.604214191436768,
      "epoch": 2.78211175616836,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042293454277291177,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18577207177877425,
      "num_tokens": 63653206.0,
      "step": 30670
    },
    {
      "entropy": 5.510841417312622,
      "epoch": 2.782565312046444,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042291016178069916,
      "loss": 5.2218,
      "mean_token_accuracy": 0.19308310151100158,
      "num_tokens": 63663427.0,
      "step": 30675
    },
    {
      "entropy": 5.529922342300415,
      "epoch": 2.7830188679245285,
      "grad_norm": 1.5,
      "learning_rate": 0.0004228857777296668,
      "loss": 5.2195,
      "mean_token_accuracy": 0.19188930094242096,
      "num_tokens": 63674269.0,
      "step": 30680
    },
    {
      "entropy": 5.614944934844971,
      "epoch": 2.7834724238026123,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004228613906203188,
      "loss": 5.233,
      "mean_token_accuracy": 0.18509361147880554,
      "num_tokens": 63683989.0,
      "step": 30685
    },
    {
      "entropy": 5.661215591430664,
      "epoch": 2.7839259796806965,
      "grad_norm": 1.609375,
      "learning_rate": 0.00042283700045315955,
      "loss": 5.2236,
      "mean_token_accuracy": 0.1848970651626587,
      "num_tokens": 63694540.0,
      "step": 30690
    },
    {
      "entropy": 5.617471504211426,
      "epoch": 2.784379535558781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042281260722869355,
      "loss": 5.2348,
      "mean_token_accuracy": 0.1874764382839203,
      "num_tokens": 63705149.0,
      "step": 30695
    },
    {
      "entropy": 5.563605928421021,
      "epoch": 2.784833091436865,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042278821094742495,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19083536118268968,
      "num_tokens": 63714061.0,
      "step": 30700
    },
    {
      "entropy": 5.673771905899048,
      "epoch": 2.7852866473149493,
      "grad_norm": 1.375,
      "learning_rate": 0.00042276381160985874,
      "loss": 5.2807,
      "mean_token_accuracy": 0.1856716513633728,
      "num_tokens": 63724732.0,
      "step": 30705
    },
    {
      "entropy": 5.622476291656494,
      "epoch": 2.7857402031930336,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004227394092164991,
      "loss": 5.2806,
      "mean_token_accuracy": 0.18530912548303605,
      "num_tokens": 63734942.0,
      "step": 30710
    },
    {
      "entropy": 5.619130373001099,
      "epoch": 2.7861937590711174,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004227150037678509,
      "loss": 5.3764,
      "mean_token_accuracy": 0.19033530950546265,
      "num_tokens": 63745326.0,
      "step": 30715
    },
    {
      "entropy": 5.663141536712646,
      "epoch": 2.7866473149492017,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004226905952644188,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18878623843193054,
      "num_tokens": 63755355.0,
      "step": 30720
    },
    {
      "entropy": 5.656942987442017,
      "epoch": 2.787100870827286,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042266618370670743,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18582262843847275,
      "num_tokens": 63765112.0,
      "step": 30725
    },
    {
      "entropy": 5.53315691947937,
      "epoch": 2.7875544267053702,
      "grad_norm": 1.71875,
      "learning_rate": 0.00042264176909522183,
      "loss": 5.1942,
      "mean_token_accuracy": 0.194913911819458,
      "num_tokens": 63774680.0,
      "step": 30730
    },
    {
      "entropy": 5.641151046752929,
      "epoch": 2.788007982583454,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004226173514304669,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19095367640256883,
      "num_tokens": 63784939.0,
      "step": 30735
    },
    {
      "entropy": 5.68661060333252,
      "epoch": 2.7884615384615383,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004225929307129474,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17308733165264129,
      "num_tokens": 63795836.0,
      "step": 30740
    },
    {
      "entropy": 5.585904693603515,
      "epoch": 2.7889150943396226,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00042256850694316844,
      "loss": 5.1916,
      "mean_token_accuracy": 0.188082917034626,
      "num_tokens": 63805607.0,
      "step": 30745
    },
    {
      "entropy": 5.563493061065674,
      "epoch": 2.789368650217707,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004225440801216353,
      "loss": 5.213,
      "mean_token_accuracy": 0.18964044451713563,
      "num_tokens": 63815081.0,
      "step": 30750
    },
    {
      "entropy": 5.504299974441528,
      "epoch": 2.789822206095791,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004225196502488529,
      "loss": 5.2158,
      "mean_token_accuracy": 0.18869644552469253,
      "num_tokens": 63825932.0,
      "step": 30755
    },
    {
      "entropy": 5.563103294372558,
      "epoch": 2.7902757619738754,
      "grad_norm": 1.6875,
      "learning_rate": 0.00042249521732532657,
      "loss": 5.1856,
      "mean_token_accuracy": 0.19121364653110504,
      "num_tokens": 63835653.0,
      "step": 30760
    },
    {
      "entropy": 5.698719215393067,
      "epoch": 2.790729317851959,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004224707813515615,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1746562361717224,
      "num_tokens": 63847657.0,
      "step": 30765
    },
    {
      "entropy": 5.695636415481568,
      "epoch": 2.7911828737300435,
      "grad_norm": 1.375,
      "learning_rate": 0.00042244634232806324,
      "loss": 5.302,
      "mean_token_accuracy": 0.1860675945878029,
      "num_tokens": 63858476.0,
      "step": 30770
    },
    {
      "entropy": 5.576734113693237,
      "epoch": 2.7916364296081277,
      "grad_norm": 1.828125,
      "learning_rate": 0.00042242190025533694,
      "loss": 5.1497,
      "mean_token_accuracy": 0.1928702101111412,
      "num_tokens": 63867970.0,
      "step": 30775
    },
    {
      "entropy": 5.650554561614991,
      "epoch": 2.792089985486212,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004223974551338881,
      "loss": 5.2242,
      "mean_token_accuracy": 0.19679431319236756,
      "num_tokens": 63878294.0,
      "step": 30780
    },
    {
      "entropy": 5.560792684555054,
      "epoch": 2.792543541364296,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004223730069642224,
      "loss": 5.2289,
      "mean_token_accuracy": 0.18628692924976348,
      "num_tokens": 63889315.0,
      "step": 30785
    },
    {
      "entropy": 5.57446985244751,
      "epoch": 2.79299709724238,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004223485557468454,
      "loss": 5.2211,
      "mean_token_accuracy": 0.1821182057261467,
      "num_tokens": 63899923.0,
      "step": 30790
    },
    {
      "entropy": 5.553267288208008,
      "epoch": 2.7934506531204644,
      "grad_norm": 2.421875,
      "learning_rate": 0.00042232410148226277,
      "loss": 5.1977,
      "mean_token_accuracy": 0.19363558292388916,
      "num_tokens": 63909813.0,
      "step": 30795
    },
    {
      "entropy": 5.5970111846923825,
      "epoch": 2.7939042089985486,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042229964417098007,
      "loss": 5.2285,
      "mean_token_accuracy": 0.18637444823980331,
      "num_tokens": 63920931.0,
      "step": 30800
    },
    {
      "entropy": 5.530473899841309,
      "epoch": 2.794357764876633,
      "grad_norm": 2.3125,
      "learning_rate": 0.00042227518381350316,
      "loss": 5.1989,
      "mean_token_accuracy": 0.19162319451570511,
      "num_tokens": 63930799.0,
      "step": 30805
    },
    {
      "entropy": 5.732982635498047,
      "epoch": 2.794811320754717,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042225072041033804,
      "loss": 5.3655,
      "mean_token_accuracy": 0.17491879910230637,
      "num_tokens": 63941763.0,
      "step": 30810
    },
    {
      "entropy": 5.709937000274659,
      "epoch": 2.7952648766328014,
      "grad_norm": 2.65625,
      "learning_rate": 0.00042222625396199043,
      "loss": 5.3383,
      "mean_token_accuracy": 0.17734934240579606,
      "num_tokens": 63952775.0,
      "step": 30815
    },
    {
      "entropy": 5.647981214523315,
      "epoch": 2.7957184325108853,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004222017844689663,
      "loss": 5.3066,
      "mean_token_accuracy": 0.18730884939432144,
      "num_tokens": 63964223.0,
      "step": 30820
    },
    {
      "entropy": 5.658669424057007,
      "epoch": 2.7961719883889695,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004221773119317719,
      "loss": 5.3534,
      "mean_token_accuracy": 0.18221426755189896,
      "num_tokens": 63976413.0,
      "step": 30825
    },
    {
      "entropy": 5.730739116668701,
      "epoch": 2.796625544267054,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042215283635091313,
      "loss": 5.3428,
      "mean_token_accuracy": 0.18465214669704438,
      "num_tokens": 63986578.0,
      "step": 30830
    },
    {
      "entropy": 5.609729480743408,
      "epoch": 2.7970791001451376,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004221283577268962,
      "loss": 5.2772,
      "mean_token_accuracy": 0.1869498983025551,
      "num_tokens": 63996856.0,
      "step": 30835
    },
    {
      "entropy": 5.5360499858856205,
      "epoch": 2.797532656023222,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00042210387606022724,
      "loss": 5.2446,
      "mean_token_accuracy": 0.1869177371263504,
      "num_tokens": 64007700.0,
      "step": 30840
    },
    {
      "entropy": 5.680777883529663,
      "epoch": 2.797986211901306,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004220793913514128,
      "loss": 5.3829,
      "mean_token_accuracy": 0.17756324857473374,
      "num_tokens": 64018771.0,
      "step": 30845
    },
    {
      "entropy": 5.645731687545776,
      "epoch": 2.7984397677793904,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004220549036009589,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18743318915367127,
      "num_tokens": 64029381.0,
      "step": 30850
    },
    {
      "entropy": 5.542285537719726,
      "epoch": 2.7988933236574747,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042203041280937224,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19256681799888611,
      "num_tokens": 64039596.0,
      "step": 30855
    },
    {
      "entropy": 5.645060873031616,
      "epoch": 2.799346879535559,
      "grad_norm": 1.5,
      "learning_rate": 0.0004220059189771592,
      "loss": 5.2702,
      "mean_token_accuracy": 0.1940206840634346,
      "num_tokens": 64049123.0,
      "step": 30860
    },
    {
      "entropy": 5.646770858764649,
      "epoch": 2.799800435413643,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042198142210482623,
      "loss": 5.3066,
      "mean_token_accuracy": 0.18473821133375168,
      "num_tokens": 64059426.0,
      "step": 30865
    },
    {
      "entropy": 5.5668456077575685,
      "epoch": 2.800253991291727,
      "grad_norm": 2.578125,
      "learning_rate": 0.00042195692219287997,
      "loss": 5.1604,
      "mean_token_accuracy": 0.19873877763748168,
      "num_tokens": 64069227.0,
      "step": 30870
    },
    {
      "entropy": 5.600939178466797,
      "epoch": 2.8007075471698113,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004219324192418272,
      "loss": 5.2672,
      "mean_token_accuracy": 0.18795639425516128,
      "num_tokens": 64078482.0,
      "step": 30875
    },
    {
      "entropy": 5.6164023876190186,
      "epoch": 2.8011611030478956,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00042190791325217447,
      "loss": 5.2287,
      "mean_token_accuracy": 0.1867414355278015,
      "num_tokens": 64088664.0,
      "step": 30880
    },
    {
      "entropy": 5.587203741073608,
      "epoch": 2.80161465892598,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004218834042244287,
      "loss": 5.1857,
      "mean_token_accuracy": 0.19113971889019013,
      "num_tokens": 64099316.0,
      "step": 30885
    },
    {
      "entropy": 5.598555612564087,
      "epoch": 2.8020682148040637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042185889215909665,
      "loss": 5.2214,
      "mean_token_accuracy": 0.18847655653953552,
      "num_tokens": 64109786.0,
      "step": 30890
    },
    {
      "entropy": 5.672716045379639,
      "epoch": 2.802521770682148,
      "grad_norm": 1.859375,
      "learning_rate": 0.00042183437705668535,
      "loss": 5.2291,
      "mean_token_accuracy": 0.18907730430364608,
      "num_tokens": 64120771.0,
      "step": 30895
    },
    {
      "entropy": 5.613127899169922,
      "epoch": 2.802975326560232,
      "grad_norm": 2.53125,
      "learning_rate": 0.00042180985891770176,
      "loss": 5.2829,
      "mean_token_accuracy": 0.18257965445518493,
      "num_tokens": 64130833.0,
      "step": 30900
    },
    {
      "entropy": 5.504565286636352,
      "epoch": 2.8034288824383164,
      "grad_norm": 2.875,
      "learning_rate": 0.0004217853377426528,
      "loss": 5.1972,
      "mean_token_accuracy": 0.18920409828424453,
      "num_tokens": 64139672.0,
      "step": 30905
    },
    {
      "entropy": 5.696752643585205,
      "epoch": 2.8038824383164007,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004217608135320456,
      "loss": 5.3906,
      "mean_token_accuracy": 0.18931836038827896,
      "num_tokens": 64150383.0,
      "step": 30910
    },
    {
      "entropy": 5.645778036117553,
      "epoch": 2.804335994194485,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004217362862863875,
      "loss": 5.207,
      "mean_token_accuracy": 0.18930382430553436,
      "num_tokens": 64160786.0,
      "step": 30915
    },
    {
      "entropy": 5.681415462493897,
      "epoch": 2.804789550072569,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042171175600618546,
      "loss": 5.31,
      "mean_token_accuracy": 0.18041132092475892,
      "num_tokens": 64172057.0,
      "step": 30920
    },
    {
      "entropy": 5.601377201080322,
      "epoch": 2.805243105950653,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042168722269194704,
      "loss": 5.2656,
      "mean_token_accuracy": 0.19044334590435028,
      "num_tokens": 64182599.0,
      "step": 30925
    },
    {
      "entropy": 5.590914487838745,
      "epoch": 2.8056966618287373,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004216626863441794,
      "loss": 5.2979,
      "mean_token_accuracy": 0.18249870985746383,
      "num_tokens": 64193356.0,
      "step": 30930
    },
    {
      "entropy": 5.670147657394409,
      "epoch": 2.8061502177068216,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042163814696339014,
      "loss": 5.3782,
      "mean_token_accuracy": 0.18140297532081603,
      "num_tokens": 64203625.0,
      "step": 30935
    },
    {
      "entropy": 5.6552574157714846,
      "epoch": 2.8066037735849054,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004216136045500866,
      "loss": 5.3616,
      "mean_token_accuracy": 0.18165597170591355,
      "num_tokens": 64214742.0,
      "step": 30940
    },
    {
      "entropy": 5.665285396575928,
      "epoch": 2.8070573294629897,
      "grad_norm": 1.625,
      "learning_rate": 0.0004215890591047763,
      "loss": 5.2815,
      "mean_token_accuracy": 0.18685321360826493,
      "num_tokens": 64223811.0,
      "step": 30945
    },
    {
      "entropy": 5.615163707733155,
      "epoch": 2.807510885341074,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042156451062796686,
      "loss": 5.168,
      "mean_token_accuracy": 0.19736808240413667,
      "num_tokens": 64234854.0,
      "step": 30950
    },
    {
      "entropy": 5.598023700714111,
      "epoch": 2.8079644412191582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042153995912016606,
      "loss": 5.2954,
      "mean_token_accuracy": 0.18301551342010497,
      "num_tokens": 64245914.0,
      "step": 30955
    },
    {
      "entropy": 5.651579236984253,
      "epoch": 2.8084179970972425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004215154045818816,
      "loss": 5.3194,
      "mean_token_accuracy": 0.18380215615034104,
      "num_tokens": 64256217.0,
      "step": 30960
    },
    {
      "entropy": 5.669658279418945,
      "epoch": 2.8088715529753268,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004214908470136212,
      "loss": 5.3235,
      "mean_token_accuracy": 0.17412110716104506,
      "num_tokens": 64267211.0,
      "step": 30965
    },
    {
      "entropy": 5.694228458404541,
      "epoch": 2.8093251088534106,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004214662864158927,
      "loss": 5.3049,
      "mean_token_accuracy": 0.18818239420652388,
      "num_tokens": 64277770.0,
      "step": 30970
    },
    {
      "entropy": 5.574645566940307,
      "epoch": 2.809778664731495,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00042144172278920413,
      "loss": 5.1944,
      "mean_token_accuracy": 0.19199759215116502,
      "num_tokens": 64287973.0,
      "step": 30975
    },
    {
      "entropy": 5.583769226074219,
      "epoch": 2.810232220609579,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004214171561340634,
      "loss": 5.2681,
      "mean_token_accuracy": 0.18070483803749085,
      "num_tokens": 64297898.0,
      "step": 30980
    },
    {
      "entropy": 5.6154717922210695,
      "epoch": 2.8106857764876634,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042139258645097863,
      "loss": 5.1565,
      "mean_token_accuracy": 0.20036027282476426,
      "num_tokens": 64308114.0,
      "step": 30985
    },
    {
      "entropy": 5.586271047592163,
      "epoch": 2.811139332365747,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00042136801374045775,
      "loss": 5.2551,
      "mean_token_accuracy": 0.1820077955722809,
      "num_tokens": 64319586.0,
      "step": 30990
    },
    {
      "entropy": 5.633994150161743,
      "epoch": 2.8115928882438315,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004213434380030091,
      "loss": 5.3199,
      "mean_token_accuracy": 0.18567479997873307,
      "num_tokens": 64330612.0,
      "step": 30995
    },
    {
      "entropy": 5.691496276855469,
      "epoch": 2.8120464441219157,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004213188592391409,
      "loss": 5.3404,
      "mean_token_accuracy": 0.17739980816841125,
      "num_tokens": 64341934.0,
      "step": 31000
    },
    {
      "entropy": 5.6041406154632565,
      "epoch": 2.8125,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004212942774493614,
      "loss": 5.1967,
      "mean_token_accuracy": 0.19006455689668655,
      "num_tokens": 64352166.0,
      "step": 31005
    },
    {
      "entropy": 5.570389127731323,
      "epoch": 2.8129535558780843,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00042126969263417885,
      "loss": 5.2614,
      "mean_token_accuracy": 0.1850027084350586,
      "num_tokens": 64361995.0,
      "step": 31010
    },
    {
      "entropy": 5.581572484970093,
      "epoch": 2.8134071117561685,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042124510479410183,
      "loss": 5.2348,
      "mean_token_accuracy": 0.18393613249063492,
      "num_tokens": 64372336.0,
      "step": 31015
    },
    {
      "entropy": 5.65903058052063,
      "epoch": 2.813860667634253,
      "grad_norm": 2.734375,
      "learning_rate": 0.00042122051392963885,
      "loss": 5.2679,
      "mean_token_accuracy": 0.1976901635527611,
      "num_tokens": 64382309.0,
      "step": 31020
    },
    {
      "entropy": 5.661507081985474,
      "epoch": 2.8143142235123366,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004211959200412984,
      "loss": 5.3243,
      "mean_token_accuracy": 0.18552620708942413,
      "num_tokens": 64393205.0,
      "step": 31025
    },
    {
      "entropy": 5.596803426742554,
      "epoch": 2.814767779390421,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004211713231295889,
      "loss": 5.2674,
      "mean_token_accuracy": 0.18969974666833878,
      "num_tokens": 64403068.0,
      "step": 31030
    },
    {
      "entropy": 5.634848165512085,
      "epoch": 2.815221335268505,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004211467231950194,
      "loss": 5.2488,
      "mean_token_accuracy": 0.18534764498472214,
      "num_tokens": 64414296.0,
      "step": 31035
    },
    {
      "entropy": 5.569733142852783,
      "epoch": 2.8156748911465894,
      "grad_norm": 1.703125,
      "learning_rate": 0.00042112212023809824,
      "loss": 5.2627,
      "mean_token_accuracy": 0.18732859343290328,
      "num_tokens": 64424748.0,
      "step": 31040
    },
    {
      "entropy": 5.629525470733642,
      "epoch": 2.8161284470246732,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004210975142593345,
      "loss": 5.1955,
      "mean_token_accuracy": 0.18666499555110933,
      "num_tokens": 64434723.0,
      "step": 31045
    },
    {
      "entropy": 5.712434339523315,
      "epoch": 2.8165820029027575,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004210729052592369,
      "loss": 5.3677,
      "mean_token_accuracy": 0.1778443470597267,
      "num_tokens": 64445864.0,
      "step": 31050
    },
    {
      "entropy": 5.538028860092163,
      "epoch": 2.8170355587808418,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004210482932383144,
      "loss": 5.2006,
      "mean_token_accuracy": 0.1837133914232254,
      "num_tokens": 64456704.0,
      "step": 31055
    },
    {
      "entropy": 5.601117897033691,
      "epoch": 2.817489114658926,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000421023678197076,
      "loss": 5.2666,
      "mean_token_accuracy": 0.18515225052833556,
      "num_tokens": 64466870.0,
      "step": 31060
    },
    {
      "entropy": 5.659518718719482,
      "epoch": 2.8179426705370103,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004209990601360307,
      "loss": 5.273,
      "mean_token_accuracy": 0.18222847878932952,
      "num_tokens": 64477069.0,
      "step": 31065
    },
    {
      "entropy": 5.629916858673096,
      "epoch": 2.8183962264150946,
      "grad_norm": 1.375,
      "learning_rate": 0.00042097443905568765,
      "loss": 5.2606,
      "mean_token_accuracy": 0.17896586805582046,
      "num_tokens": 64487165.0,
      "step": 31070
    },
    {
      "entropy": 5.596044492721558,
      "epoch": 2.8188497822931784,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00042094981495655596,
      "loss": 5.1423,
      "mean_token_accuracy": 0.19491327106952666,
      "num_tokens": 64497977.0,
      "step": 31075
    },
    {
      "entropy": 5.646826362609863,
      "epoch": 2.8193033381712627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000420925187839145,
      "loss": 5.2815,
      "mean_token_accuracy": 0.18837113976478576,
      "num_tokens": 64508767.0,
      "step": 31080
    },
    {
      "entropy": 5.588019514083863,
      "epoch": 2.819756894049347,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042090055770396383,
      "loss": 5.1692,
      "mean_token_accuracy": 0.18790753781795502,
      "num_tokens": 64519040.0,
      "step": 31085
    },
    {
      "entropy": 5.5372484683990475,
      "epoch": 2.820210449927431,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042087592455152206,
      "loss": 5.2179,
      "mean_token_accuracy": 0.19354996234178543,
      "num_tokens": 64529721.0,
      "step": 31090
    },
    {
      "entropy": 5.5735790729522705,
      "epoch": 2.820664005805515,
      "grad_norm": 1.78125,
      "learning_rate": 0.000420851288382329,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17234592288732528,
      "num_tokens": 64541299.0,
      "step": 31095
    },
    {
      "entropy": 5.682104110717773,
      "epoch": 2.8211175616835993,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042082664919689406,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19022984206676483,
      "num_tokens": 64550831.0,
      "step": 31100
    },
    {
      "entropy": 5.636130142211914,
      "epoch": 2.8215711175616836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004208020069957269,
      "loss": 5.2403,
      "mean_token_accuracy": 0.18178982138633729,
      "num_tokens": 64562076.0,
      "step": 31105
    },
    {
      "entropy": 5.531754970550537,
      "epoch": 2.822024673439768,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000420777361779337,
      "loss": 5.2435,
      "mean_token_accuracy": 0.18672736287117003,
      "num_tokens": 64572955.0,
      "step": 31110
    },
    {
      "entropy": 5.668258142471314,
      "epoch": 2.822478229317852,
      "grad_norm": 1.78125,
      "learning_rate": 0.00042075271354823427,
      "loss": 5.2937,
      "mean_token_accuracy": 0.18803591579198836,
      "num_tokens": 64584376.0,
      "step": 31115
    },
    {
      "entropy": 5.715444707870484,
      "epoch": 2.8229317851959363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004207280623029281,
      "loss": 5.2863,
      "mean_token_accuracy": 0.18319897204637528,
      "num_tokens": 64594522.0,
      "step": 31120
    },
    {
      "entropy": 5.5965252876281735,
      "epoch": 2.82338534107402,
      "grad_norm": 1.703125,
      "learning_rate": 0.00042070340804392853,
      "loss": 5.3105,
      "mean_token_accuracy": 0.18656119406223298,
      "num_tokens": 64604625.0,
      "step": 31125
    },
    {
      "entropy": 5.591050338745117,
      "epoch": 2.8238388969521044,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004206787507717454,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1849336415529251,
      "num_tokens": 64614184.0,
      "step": 31130
    },
    {
      "entropy": 5.671753835678101,
      "epoch": 2.8242924528301887,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042065409048688856,
      "loss": 5.3184,
      "mean_token_accuracy": 0.1807374283671379,
      "num_tokens": 64623019.0,
      "step": 31135
    },
    {
      "entropy": 5.630746746063233,
      "epoch": 2.824746008708273,
      "grad_norm": 1.671875,
      "learning_rate": 0.000420629427189868,
      "loss": 5.2546,
      "mean_token_accuracy": 0.18150849193334578,
      "num_tokens": 64634017.0,
      "step": 31140
    },
    {
      "entropy": 5.601252269744873,
      "epoch": 2.825199564586357,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00042060476088119376,
      "loss": 5.2714,
      "mean_token_accuracy": 0.1891172334551811,
      "num_tokens": 64645007.0,
      "step": 31145
    },
    {
      "entropy": 5.653830575942993,
      "epoch": 2.825653120464441,
      "grad_norm": 1.21875,
      "learning_rate": 0.000420580091561376,
      "loss": 5.293,
      "mean_token_accuracy": 0.18232832998037338,
      "num_tokens": 64655705.0,
      "step": 31150
    },
    {
      "entropy": 5.724048805236817,
      "epoch": 2.8261066763425253,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00042055541923092485,
      "loss": 5.3516,
      "mean_token_accuracy": 0.18079562485218048,
      "num_tokens": 64666094.0,
      "step": 31155
    },
    {
      "entropy": 5.651413869857788,
      "epoch": 2.8265602322206096,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004205307438903505,
      "loss": 5.3396,
      "mean_token_accuracy": 0.179804427921772,
      "num_tokens": 64676107.0,
      "step": 31160
    },
    {
      "entropy": 5.567063093185425,
      "epoch": 2.827013788098694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004205060655401633,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18647819757461548,
      "num_tokens": 64685846.0,
      "step": 31165
    },
    {
      "entropy": 5.58662600517273,
      "epoch": 2.827467343976778,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004204813841808736,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18775610774755477,
      "num_tokens": 64695959.0,
      "step": 31170
    },
    {
      "entropy": 5.625708866119385,
      "epoch": 2.8279208998548624,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004204566998129917,
      "loss": 5.2377,
      "mean_token_accuracy": 0.19002408534288406,
      "num_tokens": 64705594.0,
      "step": 31175
    },
    {
      "entropy": 5.632158470153809,
      "epoch": 2.828374455732946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042043201243702834,
      "loss": 5.2871,
      "mean_token_accuracy": 0.18275066912174226,
      "num_tokens": 64715974.0,
      "step": 31180
    },
    {
      "entropy": 5.575185394287109,
      "epoch": 2.8288280116110305,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004204073220534938,
      "loss": 5.1905,
      "mean_token_accuracy": 0.19203239381313325,
      "num_tokens": 64725846.0,
      "step": 31185
    },
    {
      "entropy": 5.588186025619507,
      "epoch": 2.8292815674891147,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004203826286628988,
      "loss": 5.1866,
      "mean_token_accuracy": 0.18973207771778106,
      "num_tokens": 64735789.0,
      "step": 31190
    },
    {
      "entropy": 5.592572975158691,
      "epoch": 2.8297351233671986,
      "grad_norm": 1.4375,
      "learning_rate": 0.000420357932265754,
      "loss": 5.1732,
      "mean_token_accuracy": 0.20416052937507628,
      "num_tokens": 64745738.0,
      "step": 31195
    },
    {
      "entropy": 5.612571048736572,
      "epoch": 2.830188679245283,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004203332328625701,
      "loss": 5.2661,
      "mean_token_accuracy": 0.1812368169426918,
      "num_tokens": 64756523.0,
      "step": 31200
    },
    {
      "entropy": 5.66404447555542,
      "epoch": 2.830642235123367,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00042030853045385796,
      "loss": 5.3425,
      "mean_token_accuracy": 0.17787729501724242,
      "num_tokens": 64767367.0,
      "step": 31205
    },
    {
      "entropy": 5.690195894241333,
      "epoch": 2.8310957910014514,
      "grad_norm": 1.84375,
      "learning_rate": 0.00042028382504012837,
      "loss": 5.2961,
      "mean_token_accuracy": 0.18313824087381364,
      "num_tokens": 64777729.0,
      "step": 31210
    },
    {
      "entropy": 5.69390902519226,
      "epoch": 2.8315493468795356,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042025911662189224,
      "loss": 5.3033,
      "mean_token_accuracy": 0.18488817363977433,
      "num_tokens": 64788223.0,
      "step": 31215
    },
    {
      "entropy": 5.554735374450684,
      "epoch": 2.83200290275762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042023440519966057,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18699423968791962,
      "num_tokens": 64797367.0,
      "step": 31220
    },
    {
      "entropy": 5.607228374481201,
      "epoch": 2.832456458635704,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042020969077394434,
      "loss": 5.2723,
      "mean_token_accuracy": 0.1873685508966446,
      "num_tokens": 64807179.0,
      "step": 31225
    },
    {
      "entropy": 5.643900585174561,
      "epoch": 2.832910014513788,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042018497334525475,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1828218698501587,
      "num_tokens": 64817276.0,
      "step": 31230
    },
    {
      "entropy": 5.58110466003418,
      "epoch": 2.8333635703918723,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004201602529141029,
      "loss": 5.2726,
      "mean_token_accuracy": 0.18641493767499923,
      "num_tokens": 64827509.0,
      "step": 31235
    },
    {
      "entropy": 5.597254991531372,
      "epoch": 2.8338171262699565,
      "grad_norm": 1.765625,
      "learning_rate": 0.00042013552948100003,
      "loss": 5.2497,
      "mean_token_accuracy": 0.18639560788869858,
      "num_tokens": 64836578.0,
      "step": 31240
    },
    {
      "entropy": 5.5898048877716064,
      "epoch": 2.834270682148041,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004201108030464574,
      "loss": 5.193,
      "mean_token_accuracy": 0.1956925868988037,
      "num_tokens": 64846518.0,
      "step": 31245
    },
    {
      "entropy": 5.661612176895142,
      "epoch": 2.8347242380261246,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042008607361098646,
      "loss": 5.2642,
      "mean_token_accuracy": 0.18833547830581665,
      "num_tokens": 64856304.0,
      "step": 31250
    },
    {
      "entropy": 5.5897928237915036,
      "epoch": 2.835177793904209,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004200613411750985,
      "loss": 5.1874,
      "mean_token_accuracy": 0.19561632573604584,
      "num_tokens": 64867381.0,
      "step": 31255
    },
    {
      "entropy": 5.6191400527954105,
      "epoch": 2.835631349782293,
      "grad_norm": 1.46875,
      "learning_rate": 0.000420036605739305,
      "loss": 5.2765,
      "mean_token_accuracy": 0.18680917620658874,
      "num_tokens": 64877306.0,
      "step": 31260
    },
    {
      "entropy": 5.491439247131348,
      "epoch": 2.8360849056603774,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004200118673041176,
      "loss": 5.2018,
      "mean_token_accuracy": 0.19203951507806777,
      "num_tokens": 64886932.0,
      "step": 31265
    },
    {
      "entropy": 5.666102600097656,
      "epoch": 2.8365384615384617,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041998712587004776,
      "loss": 5.2854,
      "mean_token_accuracy": 0.18015644401311875,
      "num_tokens": 64897387.0,
      "step": 31270
    },
    {
      "entropy": 5.676491975784302,
      "epoch": 2.836992017416546,
      "grad_norm": 2.46875,
      "learning_rate": 0.00041996238143760726,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18911340534687043,
      "num_tokens": 64907251.0,
      "step": 31275
    },
    {
      "entropy": 5.578410720825195,
      "epoch": 2.8374455732946298,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00041993763400730776,
      "loss": 5.245,
      "mean_token_accuracy": 0.1894570380449295,
      "num_tokens": 64917300.0,
      "step": 31280
    },
    {
      "entropy": 5.58416519165039,
      "epoch": 2.837899129172714,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004199128835796611,
      "loss": 5.1989,
      "mean_token_accuracy": 0.19048409163951874,
      "num_tokens": 64927133.0,
      "step": 31285
    },
    {
      "entropy": 5.6287007331848145,
      "epoch": 2.8383526850507983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000419888130155179,
      "loss": 5.245,
      "mean_token_accuracy": 0.18696834444999694,
      "num_tokens": 64937789.0,
      "step": 31290
    },
    {
      "entropy": 5.635310316085816,
      "epoch": 2.8388062409288826,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004198633737343735,
      "loss": 5.2886,
      "mean_token_accuracy": 0.1890277162194252,
      "num_tokens": 64948127.0,
      "step": 31295
    },
    {
      "entropy": 5.652936267852783,
      "epoch": 2.8392597968069664,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004198386143177565,
      "loss": 5.327,
      "mean_token_accuracy": 0.18450034260749817,
      "num_tokens": 64959632.0,
      "step": 31300
    },
    {
      "entropy": 5.504719018936157,
      "epoch": 2.8397133526850507,
      "grad_norm": 1.65625,
      "learning_rate": 0.00041981385190584003,
      "loss": 5.0902,
      "mean_token_accuracy": 0.19596547931432723,
      "num_tokens": 64970411.0,
      "step": 31305
    },
    {
      "entropy": 5.556020069122314,
      "epoch": 2.840166908563135,
      "grad_norm": 1.375,
      "learning_rate": 0.0004197890864991362,
      "loss": 5.2241,
      "mean_token_accuracy": 0.18508636355400085,
      "num_tokens": 64980748.0,
      "step": 31310
    },
    {
      "entropy": 5.596821975708008,
      "epoch": 2.840620464441219,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004197643180981572,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18818315416574477,
      "num_tokens": 64990818.0,
      "step": 31315
    },
    {
      "entropy": 5.655906534194946,
      "epoch": 2.8410740203193035,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004197395467034152,
      "loss": 5.2329,
      "mean_token_accuracy": 0.19212736040353776,
      "num_tokens": 65001993.0,
      "step": 31320
    },
    {
      "entropy": 5.539413166046143,
      "epoch": 2.8415275761973877,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004197147723154225,
      "loss": 5.2129,
      "mean_token_accuracy": 0.19604113548994065,
      "num_tokens": 65013007.0,
      "step": 31325
    },
    {
      "entropy": 5.679647397994995,
      "epoch": 2.8419811320754715,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004196899949346914,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18667927980422974,
      "num_tokens": 65023595.0,
      "step": 31330
    },
    {
      "entropy": 5.60546612739563,
      "epoch": 2.842434687953556,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004196652145617343,
      "loss": 5.2321,
      "mean_token_accuracy": 0.1882070779800415,
      "num_tokens": 65033631.0,
      "step": 31335
    },
    {
      "entropy": 5.648716068267822,
      "epoch": 2.84288824383164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004196404311970638,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18758751600980758,
      "num_tokens": 65044059.0,
      "step": 31340
    },
    {
      "entropy": 5.596531629562378,
      "epoch": 2.8433417997097243,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004196156448411922,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18960454910993577,
      "num_tokens": 65053931.0,
      "step": 31345
    },
    {
      "entropy": 5.598639345169067,
      "epoch": 2.843795355587808,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041959085549463234,
      "loss": 5.2347,
      "mean_token_accuracy": 0.19547007828950883,
      "num_tokens": 65064011.0,
      "step": 31350
    },
    {
      "entropy": 5.656743001937866,
      "epoch": 2.8442489114658924,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004195660631578966,
      "loss": 5.3312,
      "mean_token_accuracy": 0.18463789522647858,
      "num_tokens": 65074417.0,
      "step": 31355
    },
    {
      "entropy": 5.56186351776123,
      "epoch": 2.8447024673439767,
      "grad_norm": 2.046875,
      "learning_rate": 0.000419541267831498,
      "loss": 5.2269,
      "mean_token_accuracy": 0.19457645416259767,
      "num_tokens": 65084593.0,
      "step": 31360
    },
    {
      "entropy": 5.596468782424926,
      "epoch": 2.845156023222061,
      "grad_norm": 1.40625,
      "learning_rate": 0.000419516469515949,
      "loss": 5.2581,
      "mean_token_accuracy": 0.18989029079675673,
      "num_tokens": 65095047.0,
      "step": 31365
    },
    {
      "entropy": 5.659003114700317,
      "epoch": 2.8456095791001452,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041949166821176264,
      "loss": 5.3427,
      "mean_token_accuracy": 0.18687236309051514,
      "num_tokens": 65104434.0,
      "step": 31370
    },
    {
      "entropy": 5.695283365249634,
      "epoch": 2.8460631349782295,
      "grad_norm": 1.640625,
      "learning_rate": 0.00041946686391945183,
      "loss": 5.3236,
      "mean_token_accuracy": 0.18786962777376176,
      "num_tokens": 65114039.0,
      "step": 31375
    },
    {
      "entropy": 5.718349599838257,
      "epoch": 2.8465166908563138,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004194420566395294,
      "loss": 5.3773,
      "mean_token_accuracy": 0.17271018624305726,
      "num_tokens": 65125387.0,
      "step": 31380
    },
    {
      "entropy": 5.677069473266601,
      "epoch": 2.8469702467343976,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004194172463725084,
      "loss": 5.2928,
      "mean_token_accuracy": 0.19216859340667725,
      "num_tokens": 65136351.0,
      "step": 31385
    },
    {
      "entropy": 5.603289270401001,
      "epoch": 2.847423802612482,
      "grad_norm": 1.671875,
      "learning_rate": 0.00041939243311890195,
      "loss": 5.2608,
      "mean_token_accuracy": 0.18359043598175048,
      "num_tokens": 65147343.0,
      "step": 31390
    },
    {
      "entropy": 5.537886095046997,
      "epoch": 2.847877358490566,
      "grad_norm": 1.25,
      "learning_rate": 0.00041936761687922324,
      "loss": 5.2075,
      "mean_token_accuracy": 0.1935802936553955,
      "num_tokens": 65157097.0,
      "step": 31395
    },
    {
      "entropy": 5.59920825958252,
      "epoch": 2.8483309143686504,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041934279765398535,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18809058219194413,
      "num_tokens": 65167546.0,
      "step": 31400
    },
    {
      "entropy": 5.645564079284668,
      "epoch": 2.848784470246734,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004193179754437017,
      "loss": 5.3023,
      "mean_token_accuracy": 0.18162939101457595,
      "num_tokens": 65178497.0,
      "step": 31405
    },
    {
      "entropy": 5.579621648788452,
      "epoch": 2.8492380261248185,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004192931502488855,
      "loss": 5.1818,
      "mean_token_accuracy": 0.1931413695216179,
      "num_tokens": 65189530.0,
      "step": 31410
    },
    {
      "entropy": 5.580009841918946,
      "epoch": 2.8496915820029027,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041926832207005025,
      "loss": 5.2219,
      "mean_token_accuracy": 0.1904563844203949,
      "num_tokens": 65199836.0,
      "step": 31415
    },
    {
      "entropy": 5.538672256469726,
      "epoch": 2.850145137880987,
      "grad_norm": 1.75,
      "learning_rate": 0.0004192434909077092,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19881729036569595,
      "num_tokens": 65210208.0,
      "step": 31420
    },
    {
      "entropy": 5.583579158782959,
      "epoch": 2.8505986937590713,
      "grad_norm": 2.03125,
      "learning_rate": 0.00041921865676237605,
      "loss": 5.2746,
      "mean_token_accuracy": 0.18647231459617614,
      "num_tokens": 65220954.0,
      "step": 31425
    },
    {
      "entropy": 5.603621578216552,
      "epoch": 2.8510522496371555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041919381963456437,
      "loss": 5.2275,
      "mean_token_accuracy": 0.19046059995889664,
      "num_tokens": 65230870.0,
      "step": 31430
    },
    {
      "entropy": 5.658203840255737,
      "epoch": 2.8515058055152394,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004191689795247877,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18771959990262985,
      "num_tokens": 65241913.0,
      "step": 31435
    },
    {
      "entropy": 5.675348997116089,
      "epoch": 2.8519593613933236,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004191441364335598,
      "loss": 5.2513,
      "mean_token_accuracy": 0.18847070038318633,
      "num_tokens": 65252513.0,
      "step": 31440
    },
    {
      "entropy": 5.5478675842285154,
      "epoch": 2.852412917271408,
      "grad_norm": 1.84375,
      "learning_rate": 0.00041911929036139457,
      "loss": 5.2484,
      "mean_token_accuracy": 0.18716566413640975,
      "num_tokens": 65262695.0,
      "step": 31445
    },
    {
      "entropy": 5.676504611968994,
      "epoch": 2.852866473149492,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00041909444130880553,
      "loss": 5.3244,
      "mean_token_accuracy": 0.18779926002025604,
      "num_tokens": 65274131.0,
      "step": 31450
    },
    {
      "entropy": 5.658473634719849,
      "epoch": 2.853320029027576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041906958927630673,
      "loss": 5.278,
      "mean_token_accuracy": 0.18793666064739228,
      "num_tokens": 65285080.0,
      "step": 31455
    },
    {
      "entropy": 5.714796829223633,
      "epoch": 2.8537735849056602,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004190447342644122,
      "loss": 5.4678,
      "mean_token_accuracy": 0.1808447241783142,
      "num_tokens": 65296512.0,
      "step": 31460
    },
    {
      "entropy": 5.580975770950317,
      "epoch": 2.8542271407837445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004190198762736358,
      "loss": 5.175,
      "mean_token_accuracy": 0.19582240283489227,
      "num_tokens": 65306054.0,
      "step": 31465
    },
    {
      "entropy": 5.54708080291748,
      "epoch": 2.8546806966618288,
      "grad_norm": 1.375,
      "learning_rate": 0.0004189950153044917,
      "loss": 5.2294,
      "mean_token_accuracy": 0.18743917644023894,
      "num_tokens": 65315046.0,
      "step": 31470
    },
    {
      "entropy": 5.593299055099488,
      "epoch": 2.855134252539913,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004189701513574939,
      "loss": 5.1874,
      "mean_token_accuracy": 0.189424504339695,
      "num_tokens": 65325482.0,
      "step": 31475
    },
    {
      "entropy": 5.610102128982544,
      "epoch": 2.8555878084179973,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004189452844331567,
      "loss": 5.1562,
      "mean_token_accuracy": 0.19020346850156783,
      "num_tokens": 65335518.0,
      "step": 31480
    },
    {
      "entropy": 5.611487770080567,
      "epoch": 2.856041364296081,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00041892041453199443,
      "loss": 5.3269,
      "mean_token_accuracy": 0.18688176870346068,
      "num_tokens": 65346079.0,
      "step": 31485
    },
    {
      "entropy": 5.572953224182129,
      "epoch": 2.8564949201741654,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004188955416545212,
      "loss": 5.205,
      "mean_token_accuracy": 0.1885611519217491,
      "num_tokens": 65356719.0,
      "step": 31490
    },
    {
      "entropy": 5.65887188911438,
      "epoch": 2.8569484760522497,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004188706658012516,
      "loss": 5.262,
      "mean_token_accuracy": 0.1825496569275856,
      "num_tokens": 65367877.0,
      "step": 31495
    },
    {
      "entropy": 5.666328239440918,
      "epoch": 2.857402031930334,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041884578697269994,
      "loss": 5.3244,
      "mean_token_accuracy": 0.1748809039592743,
      "num_tokens": 65379624.0,
      "step": 31500
    },
    {
      "entropy": 5.660658502578736,
      "epoch": 2.8578555878084178,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004188209051693807,
      "loss": 5.3106,
      "mean_token_accuracy": 0.18833909332752227,
      "num_tokens": 65390417.0,
      "step": 31505
    },
    {
      "entropy": 5.482662343978882,
      "epoch": 2.858309143686502,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00041879602039180856,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19441743195056915,
      "num_tokens": 65401681.0,
      "step": 31510
    },
    {
      "entropy": 5.566654109954834,
      "epoch": 2.8587626995645863,
      "grad_norm": 1.546875,
      "learning_rate": 0.00041877113264049805,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18757978677749634,
      "num_tokens": 65412520.0,
      "step": 31515
    },
    {
      "entropy": 5.6101678848266605,
      "epoch": 2.8592162554426706,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041874624191596383,
      "loss": 5.2537,
      "mean_token_accuracy": 0.18909413665533065,
      "num_tokens": 65423201.0,
      "step": 31520
    },
    {
      "entropy": 5.5729334354400635,
      "epoch": 2.859669811320755,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004187213482187207,
      "loss": 5.2522,
      "mean_token_accuracy": 0.1845548063516617,
      "num_tokens": 65432778.0,
      "step": 31525
    },
    {
      "entropy": 5.689258337020874,
      "epoch": 2.860123367198839,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004186964515492835,
      "loss": 5.2724,
      "mean_token_accuracy": 0.19086347818374633,
      "num_tokens": 65442323.0,
      "step": 31530
    },
    {
      "entropy": 5.58055157661438,
      "epoch": 2.8605769230769234,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004186715519081671,
      "loss": 5.2322,
      "mean_token_accuracy": 0.18872081637382507,
      "num_tokens": 65451980.0,
      "step": 31535
    },
    {
      "entropy": 5.539878034591675,
      "epoch": 2.861030478955007,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004186466492958864,
      "loss": 5.1923,
      "mean_token_accuracy": 0.1936275914311409,
      "num_tokens": 65463359.0,
      "step": 31540
    },
    {
      "entropy": 5.6285741329193115,
      "epoch": 2.8614840348330914,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004186217437129562,
      "loss": 5.2684,
      "mean_token_accuracy": 0.18675801753997803,
      "num_tokens": 65474155.0,
      "step": 31545
    },
    {
      "entropy": 5.618401384353637,
      "epoch": 2.8619375907111757,
      "grad_norm": 1.875,
      "learning_rate": 0.00041859683515989196,
      "loss": 5.1965,
      "mean_token_accuracy": 0.19436110705137252,
      "num_tokens": 65484884.0,
      "step": 31550
    },
    {
      "entropy": 5.573398065567017,
      "epoch": 2.8623911465892595,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00041857192363720845,
      "loss": 5.266,
      "mean_token_accuracy": 0.18864387571811675,
      "num_tokens": 65494380.0,
      "step": 31555
    },
    {
      "entropy": 5.597336530685425,
      "epoch": 2.862844702467344,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004185470091454211,
      "loss": 5.245,
      "mean_token_accuracy": 0.18858703523874282,
      "num_tokens": 65505735.0,
      "step": 31560
    },
    {
      "entropy": 5.66780948638916,
      "epoch": 2.863298258345428,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004185220916850449,
      "loss": 5.1979,
      "mean_token_accuracy": 0.18826186507940293,
      "num_tokens": 65515956.0,
      "step": 31565
    },
    {
      "entropy": 5.514501714706421,
      "epoch": 2.8637518142235123,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041849717125659537,
      "loss": 5.1737,
      "mean_token_accuracy": 0.19076795279979705,
      "num_tokens": 65526369.0,
      "step": 31570
    },
    {
      "entropy": 5.553296279907227,
      "epoch": 2.8642053701015966,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004184722478605877,
      "loss": 5.21,
      "mean_token_accuracy": 0.19230148941278458,
      "num_tokens": 65536032.0,
      "step": 31575
    },
    {
      "entropy": 5.551990222930908,
      "epoch": 2.864658925979681,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004184473214975374,
      "loss": 5.2554,
      "mean_token_accuracy": 0.18478586077690123,
      "num_tokens": 65546121.0,
      "step": 31580
    },
    {
      "entropy": 5.611573266983032,
      "epoch": 2.865112481857765,
      "grad_norm": 3.28125,
      "learning_rate": 0.00041842239216795994,
      "loss": 5.2093,
      "mean_token_accuracy": 0.18939178436994553,
      "num_tokens": 65555915.0,
      "step": 31585
    },
    {
      "entropy": 5.600993728637695,
      "epoch": 2.865566037735849,
      "grad_norm": 1.5625,
      "learning_rate": 0.00041839745987237086,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1877207338809967,
      "num_tokens": 65566267.0,
      "step": 31590
    },
    {
      "entropy": 5.629128646850586,
      "epoch": 2.866019593613933,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004183725246112857,
      "loss": 5.2553,
      "mean_token_accuracy": 0.195936980843544,
      "num_tokens": 65577126.0,
      "step": 31595
    },
    {
      "entropy": 5.640489292144776,
      "epoch": 2.8664731494920175,
      "grad_norm": 1.625,
      "learning_rate": 0.00041834758638522034,
      "loss": 5.3585,
      "mean_token_accuracy": 0.17603804767131806,
      "num_tokens": 65587158.0,
      "step": 31600
    },
    {
      "entropy": 5.57752947807312,
      "epoch": 2.8669267053701017,
      "grad_norm": 2.5625,
      "learning_rate": 0.00041832264519469026,
      "loss": 5.2352,
      "mean_token_accuracy": 0.18610770106315613,
      "num_tokens": 65597720.0,
      "step": 31605
    },
    {
      "entropy": 5.660631895065308,
      "epoch": 2.8673802612481856,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041829770104021143,
      "loss": 5.2659,
      "mean_token_accuracy": 0.1883737251162529,
      "num_tokens": 65607409.0,
      "step": 31610
    },
    {
      "entropy": 5.59232907295227,
      "epoch": 2.86783381712627,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041827275392229956,
      "loss": 5.2085,
      "mean_token_accuracy": 0.19769665598869324,
      "num_tokens": 65616785.0,
      "step": 31615
    },
    {
      "entropy": 5.586869096755981,
      "epoch": 2.868287373004354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004182478038414706,
      "loss": 5.3122,
      "mean_token_accuracy": 0.18615957647562026,
      "num_tokens": 65625794.0,
      "step": 31620
    },
    {
      "entropy": 5.530925035476685,
      "epoch": 2.8687409288824384,
      "grad_norm": 1.5625,
      "learning_rate": 0.00041822285079824056,
      "loss": 5.1277,
      "mean_token_accuracy": 0.1928309366106987,
      "num_tokens": 65635789.0,
      "step": 31625
    },
    {
      "entropy": 5.6708354473114015,
      "epoch": 2.8691944847605226,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004181978947931256,
      "loss": 5.3632,
      "mean_token_accuracy": 0.1840444788336754,
      "num_tokens": 65646266.0,
      "step": 31630
    },
    {
      "entropy": 5.599059677124023,
      "epoch": 2.869648040638607,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004181729358266415,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18551996052265168,
      "num_tokens": 65656538.0,
      "step": 31635
    },
    {
      "entropy": 5.674583578109742,
      "epoch": 2.8701015965166907,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00041814797389930474,
      "loss": 5.2716,
      "mean_token_accuracy": 0.18843873739242553,
      "num_tokens": 65666926.0,
      "step": 31640
    },
    {
      "entropy": 5.594834518432617,
      "epoch": 2.870555152394775,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004181230090116313,
      "loss": 5.2181,
      "mean_token_accuracy": 0.1853316217660904,
      "num_tokens": 65677219.0,
      "step": 31645
    },
    {
      "entropy": 5.577638101577759,
      "epoch": 2.8710087082728593,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00041809804116413763,
      "loss": 5.2981,
      "mean_token_accuracy": 0.17881433814764022,
      "num_tokens": 65688281.0,
      "step": 31650
    },
    {
      "entropy": 5.738239002227783,
      "epoch": 2.8714622641509435,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041807307035734,
      "loss": 5.3553,
      "mean_token_accuracy": 0.1788487032055855,
      "num_tokens": 65698815.0,
      "step": 31655
    },
    {
      "entropy": 5.630369853973389,
      "epoch": 2.8719158200290273,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004180480965917548,
      "loss": 5.2861,
      "mean_token_accuracy": 0.18564891517162324,
      "num_tokens": 65709481.0,
      "step": 31660
    },
    {
      "entropy": 5.63307557106018,
      "epoch": 2.8723693759071116,
      "grad_norm": 1.859375,
      "learning_rate": 0.00041802311986789854,
      "loss": 5.2453,
      "mean_token_accuracy": 0.1818622887134552,
      "num_tokens": 65720561.0,
      "step": 31665
    },
    {
      "entropy": 5.621208763122558,
      "epoch": 2.872822931785196,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041799814018628765,
      "loss": 5.2869,
      "mean_token_accuracy": 0.17980922311544417,
      "num_tokens": 65730384.0,
      "step": 31670
    },
    {
      "entropy": 5.593001794815064,
      "epoch": 2.87327648766328,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041797315754743883,
      "loss": 5.2629,
      "mean_token_accuracy": 0.1871709942817688,
      "num_tokens": 65740711.0,
      "step": 31675
    },
    {
      "entropy": 5.751163005828857,
      "epoch": 2.8737300435413644,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041794817195186867,
      "loss": 5.3651,
      "mean_token_accuracy": 0.18084964156150818,
      "num_tokens": 65751033.0,
      "step": 31680
    },
    {
      "entropy": 5.667291259765625,
      "epoch": 2.8741835994194487,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004179231834000939,
      "loss": 5.3289,
      "mean_token_accuracy": 0.17802528142929078,
      "num_tokens": 65761500.0,
      "step": 31685
    },
    {
      "entropy": 5.503406524658203,
      "epoch": 2.8746371552975325,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041789819189263117,
      "loss": 5.1566,
      "mean_token_accuracy": 0.19512635022401809,
      "num_tokens": 65771095.0,
      "step": 31690
    },
    {
      "entropy": 5.59419093132019,
      "epoch": 2.8750907111756168,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041787319742999756,
      "loss": 5.2416,
      "mean_token_accuracy": 0.19277824759483336,
      "num_tokens": 65782058.0,
      "step": 31695
    },
    {
      "entropy": 5.61789984703064,
      "epoch": 2.875544267053701,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004178482000127098,
      "loss": 5.2755,
      "mean_token_accuracy": 0.1920984148979187,
      "num_tokens": 65791710.0,
      "step": 31700
    },
    {
      "entropy": 5.603828001022339,
      "epoch": 2.8759978229317853,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004178231996412848,
      "loss": 5.3204,
      "mean_token_accuracy": 0.1777525708079338,
      "num_tokens": 65802180.0,
      "step": 31705
    },
    {
      "entropy": 5.575703859329224,
      "epoch": 2.876451378809869,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004177981963162396,
      "loss": 5.185,
      "mean_token_accuracy": 0.19526736438274384,
      "num_tokens": 65812998.0,
      "step": 31710
    },
    {
      "entropy": 5.639235258102417,
      "epoch": 2.8769049346879534,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004177731900380914,
      "loss": 5.2788,
      "mean_token_accuracy": 0.18588760793209075,
      "num_tokens": 65824210.0,
      "step": 31715
    },
    {
      "entropy": 5.5763157367706295,
      "epoch": 2.8773584905660377,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004177481808073572,
      "loss": 5.2221,
      "mean_token_accuracy": 0.19148176163434982,
      "num_tokens": 65834322.0,
      "step": 31720
    },
    {
      "entropy": 5.599327516555786,
      "epoch": 2.877812046444122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041772316862455426,
      "loss": 5.1954,
      "mean_token_accuracy": 0.1930982619524002,
      "num_tokens": 65845347.0,
      "step": 31725
    },
    {
      "entropy": 5.635932731628418,
      "epoch": 2.878265602322206,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041769815349019986,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18929768055677415,
      "num_tokens": 65855616.0,
      "step": 31730
    },
    {
      "entropy": 5.607989120483398,
      "epoch": 2.8787191582002905,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041767313540481125,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18834032565355302,
      "num_tokens": 65865542.0,
      "step": 31735
    },
    {
      "entropy": 5.555176877975464,
      "epoch": 2.8791727140783747,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004176481143689058,
      "loss": 5.206,
      "mean_token_accuracy": 0.18917035162448884,
      "num_tokens": 65874478.0,
      "step": 31740
    },
    {
      "entropy": 5.6582857131958,
      "epoch": 2.8796262699564585,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004176230903830011,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1889253929257393,
      "num_tokens": 65884609.0,
      "step": 31745
    },
    {
      "entropy": 5.5851034164428714,
      "epoch": 2.880079825834543,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004175980634476145,
      "loss": 5.2388,
      "mean_token_accuracy": 0.19385414719581603,
      "num_tokens": 65894486.0,
      "step": 31750
    },
    {
      "entropy": 5.534439420700073,
      "epoch": 2.880533381712627,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004175730335632635,
      "loss": 5.1894,
      "mean_token_accuracy": 0.19007062166929245,
      "num_tokens": 65904643.0,
      "step": 31755
    },
    {
      "entropy": 5.50881667137146,
      "epoch": 2.8809869375907113,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00041754800073046593,
      "loss": 5.2001,
      "mean_token_accuracy": 0.1956702083349228,
      "num_tokens": 65914107.0,
      "step": 31760
    },
    {
      "entropy": 5.609297895431519,
      "epoch": 2.881440493468795,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004175229649497394,
      "loss": 5.1889,
      "mean_token_accuracy": 0.1857231006026268,
      "num_tokens": 65925112.0,
      "step": 31765
    },
    {
      "entropy": 5.660555648803711,
      "epoch": 2.8818940493468794,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004174979262216016,
      "loss": 5.2669,
      "mean_token_accuracy": 0.18322535008192062,
      "num_tokens": 65934624.0,
      "step": 31770
    },
    {
      "entropy": 5.618071746826172,
      "epoch": 2.8823476052249637,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004174728845465704,
      "loss": 5.266,
      "mean_token_accuracy": 0.1807321384549141,
      "num_tokens": 65945951.0,
      "step": 31775
    },
    {
      "entropy": 5.5602569580078125,
      "epoch": 2.882801161103048,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004174478399251636,
      "loss": 5.1601,
      "mean_token_accuracy": 0.19336748719215394,
      "num_tokens": 65956871.0,
      "step": 31780
    },
    {
      "entropy": 5.6783233165740965,
      "epoch": 2.8832547169811322,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004174227923578992,
      "loss": 5.3818,
      "mean_token_accuracy": 0.18175965249538423,
      "num_tokens": 65967814.0,
      "step": 31785
    },
    {
      "entropy": 5.613682746887207,
      "epoch": 2.8837082728592165,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004173977418452952,
      "loss": 5.2683,
      "mean_token_accuracy": 0.18058339059352874,
      "num_tokens": 65978118.0,
      "step": 31790
    },
    {
      "entropy": 5.5702338218688965,
      "epoch": 2.8841618287373003,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041737268838786955,
      "loss": 5.2015,
      "mean_token_accuracy": 0.1920098379254341,
      "num_tokens": 65988240.0,
      "step": 31795
    },
    {
      "entropy": 5.618526601791382,
      "epoch": 2.8846153846153846,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004173476319861404,
      "loss": 5.1884,
      "mean_token_accuracy": 0.1912033289670944,
      "num_tokens": 65998126.0,
      "step": 31800
    },
    {
      "entropy": 5.605703544616699,
      "epoch": 2.885068940493469,
      "grad_norm": 1.375,
      "learning_rate": 0.0004173225726406261,
      "loss": 5.2621,
      "mean_token_accuracy": 0.1894989416003227,
      "num_tokens": 66009118.0,
      "step": 31805
    },
    {
      "entropy": 5.548100900650025,
      "epoch": 2.885522496371553,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041729751035184457,
      "loss": 5.2964,
      "mean_token_accuracy": 0.1878051206469536,
      "num_tokens": 66019340.0,
      "step": 31810
    },
    {
      "entropy": 5.576782703399658,
      "epoch": 2.885976052249637,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041727244512031425,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1913926973938942,
      "num_tokens": 66029784.0,
      "step": 31815
    },
    {
      "entropy": 5.670319747924805,
      "epoch": 2.886429608127721,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004172473769465536,
      "loss": 5.3606,
      "mean_token_accuracy": 0.1773361951112747,
      "num_tokens": 66039987.0,
      "step": 31820
    },
    {
      "entropy": 5.638870573043823,
      "epoch": 2.8868831640058055,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00041722230583108097,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18818138539791107,
      "num_tokens": 66049150.0,
      "step": 31825
    },
    {
      "entropy": 5.600475406646728,
      "epoch": 2.8873367198838897,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041719723177441467,
      "loss": 5.3029,
      "mean_token_accuracy": 0.18732334077358245,
      "num_tokens": 66060719.0,
      "step": 31830
    },
    {
      "entropy": 5.630538749694824,
      "epoch": 2.887790275761974,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004171721547770735,
      "loss": 5.2817,
      "mean_token_accuracy": 0.18529011607170104,
      "num_tokens": 66071126.0,
      "step": 31835
    },
    {
      "entropy": 5.64294843673706,
      "epoch": 2.8882438316400583,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004171470748395759,
      "loss": 5.1917,
      "mean_token_accuracy": 0.19330438375473022,
      "num_tokens": 66080928.0,
      "step": 31840
    },
    {
      "entropy": 5.606819438934326,
      "epoch": 2.888697387518142,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004171219919624406,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18455016016960143,
      "num_tokens": 66090900.0,
      "step": 31845
    },
    {
      "entropy": 5.610186338424683,
      "epoch": 2.8891509433962264,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041709690614618624,
      "loss": 5.2636,
      "mean_token_accuracy": 0.18702971041202546,
      "num_tokens": 66100720.0,
      "step": 31850
    },
    {
      "entropy": 5.622090291976929,
      "epoch": 2.8896044992743106,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041707181739133163,
      "loss": 5.261,
      "mean_token_accuracy": 0.1802691176533699,
      "num_tokens": 66110780.0,
      "step": 31855
    },
    {
      "entropy": 5.608022546768188,
      "epoch": 2.890058055152395,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004170467256983957,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1975102886557579,
      "num_tokens": 66120976.0,
      "step": 31860
    },
    {
      "entropy": 5.5876133918762205,
      "epoch": 2.8905116110304787,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004170216310678973,
      "loss": 5.239,
      "mean_token_accuracy": 0.19139662086963655,
      "num_tokens": 66130221.0,
      "step": 31865
    },
    {
      "entropy": 5.554192590713501,
      "epoch": 2.890965166908563,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041699653350035527,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19689172357320786,
      "num_tokens": 66140060.0,
      "step": 31870
    },
    {
      "entropy": 5.6120298385620115,
      "epoch": 2.8914187227866472,
      "grad_norm": 1.59375,
      "learning_rate": 0.00041697143299628886,
      "loss": 5.2621,
      "mean_token_accuracy": 0.19115497916936874,
      "num_tokens": 66151640.0,
      "step": 31875
    },
    {
      "entropy": 5.6312172412872314,
      "epoch": 2.8918722786647315,
      "grad_norm": 1.78125,
      "learning_rate": 0.00041694632955621697,
      "loss": 5.2364,
      "mean_token_accuracy": 0.19209470450878144,
      "num_tokens": 66162923.0,
      "step": 31880
    },
    {
      "entropy": 5.664022445678711,
      "epoch": 2.8923258345428158,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041692122318065885,
      "loss": 5.2908,
      "mean_token_accuracy": 0.18795607835054398,
      "num_tokens": 66173108.0,
      "step": 31885
    },
    {
      "entropy": 5.573760223388672,
      "epoch": 2.8927793904209,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041689611387013363,
      "loss": 5.202,
      "mean_token_accuracy": 0.19161785989999772,
      "num_tokens": 66185280.0,
      "step": 31890
    },
    {
      "entropy": 5.556868362426758,
      "epoch": 2.8932329462989843,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004168710016251606,
      "loss": 5.2157,
      "mean_token_accuracy": 0.1858002468943596,
      "num_tokens": 66196020.0,
      "step": 31895
    },
    {
      "entropy": 5.538808917999267,
      "epoch": 2.893686502177068,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004168458864462592,
      "loss": 5.1544,
      "mean_token_accuracy": 0.18941329270601273,
      "num_tokens": 66205753.0,
      "step": 31900
    },
    {
      "entropy": 5.671002388000488,
      "epoch": 2.8941400580551524,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041682076833394864,
      "loss": 5.3474,
      "mean_token_accuracy": 0.18281810432672502,
      "num_tokens": 66218415.0,
      "step": 31905
    },
    {
      "entropy": 5.640676546096802,
      "epoch": 2.8945936139332367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041679564728874845,
      "loss": 5.3235,
      "mean_token_accuracy": 0.18322136998176575,
      "num_tokens": 66228090.0,
      "step": 31910
    },
    {
      "entropy": 5.567836809158325,
      "epoch": 2.8950471698113205,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004167705233111782,
      "loss": 5.1646,
      "mean_token_accuracy": 0.19285901486873627,
      "num_tokens": 66237587.0,
      "step": 31915
    },
    {
      "entropy": 5.5972044467926025,
      "epoch": 2.8955007256894048,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00041674539640175733,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18657470345497132,
      "num_tokens": 66248162.0,
      "step": 31920
    },
    {
      "entropy": 5.562894821166992,
      "epoch": 2.895954281567489,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004167202665610055,
      "loss": 5.184,
      "mean_token_accuracy": 0.19470485746860505,
      "num_tokens": 66258370.0,
      "step": 31925
    },
    {
      "entropy": 5.618650674819946,
      "epoch": 2.8964078374455733,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00041669513378944254,
      "loss": 5.2641,
      "mean_token_accuracy": 0.1872525081038475,
      "num_tokens": 66269267.0,
      "step": 31930
    },
    {
      "entropy": 5.681403779983521,
      "epoch": 2.8968613933236576,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041666999808758815,
      "loss": 5.2702,
      "mean_token_accuracy": 0.18616410195827485,
      "num_tokens": 66280077.0,
      "step": 31935
    },
    {
      "entropy": 5.584911441802978,
      "epoch": 2.897314949201742,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000416644859455962,
      "loss": 5.2204,
      "mean_token_accuracy": 0.19150887578725814,
      "num_tokens": 66291678.0,
      "step": 31940
    },
    {
      "entropy": 5.606530666351318,
      "epoch": 2.897768505079826,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004166197178950841,
      "loss": 5.2161,
      "mean_token_accuracy": 0.18905897438526154,
      "num_tokens": 66302046.0,
      "step": 31945
    },
    {
      "entropy": 5.582103633880616,
      "epoch": 2.89822206095791,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041659457340547436,
      "loss": 5.2336,
      "mean_token_accuracy": 0.18427304178476334,
      "num_tokens": 66311888.0,
      "step": 31950
    },
    {
      "entropy": 5.54994306564331,
      "epoch": 2.898675616835994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041656942598765275,
      "loss": 5.2055,
      "mean_token_accuracy": 0.19233627170324324,
      "num_tokens": 66321507.0,
      "step": 31955
    },
    {
      "entropy": 5.523421335220337,
      "epoch": 2.8991291727140784,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00041654427564213924,
      "loss": 5.1987,
      "mean_token_accuracy": 0.19013579189777374,
      "num_tokens": 66331458.0,
      "step": 31960
    },
    {
      "entropy": 5.596927547454834,
      "epoch": 2.8995827285921627,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00041651912236945416,
      "loss": 5.2172,
      "mean_token_accuracy": 0.1908506378531456,
      "num_tokens": 66340875.0,
      "step": 31965
    },
    {
      "entropy": 5.595533323287964,
      "epoch": 2.9000362844702465,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00041649396617011753,
      "loss": 5.2877,
      "mean_token_accuracy": 0.18475213646888733,
      "num_tokens": 66350784.0,
      "step": 31970
    },
    {
      "entropy": 5.610494184494018,
      "epoch": 2.900489840348331,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004164688070446496,
      "loss": 5.3012,
      "mean_token_accuracy": 0.18238919228315353,
      "num_tokens": 66361591.0,
      "step": 31975
    },
    {
      "entropy": 5.625517797470093,
      "epoch": 2.900943396226415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004164436449935707,
      "loss": 5.1997,
      "mean_token_accuracy": 0.19089362919330596,
      "num_tokens": 66371540.0,
      "step": 31980
    },
    {
      "entropy": 5.627355289459229,
      "epoch": 2.9013969521044993,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041641848001740126,
      "loss": 5.2573,
      "mean_token_accuracy": 0.1904330372810364,
      "num_tokens": 66382053.0,
      "step": 31985
    },
    {
      "entropy": 5.548773002624512,
      "epoch": 2.9018505079825836,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004163933121166615,
      "loss": 5.2057,
      "mean_token_accuracy": 0.19480858147144317,
      "num_tokens": 66392563.0,
      "step": 31990
    },
    {
      "entropy": 5.595341062545776,
      "epoch": 2.902304063860668,
      "grad_norm": 1.59375,
      "learning_rate": 0.000416368141291872,
      "loss": 5.175,
      "mean_token_accuracy": 0.18991029262542725,
      "num_tokens": 66403582.0,
      "step": 31995
    },
    {
      "entropy": 5.52885332107544,
      "epoch": 2.9027576197387517,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041634296754355346,
      "loss": 5.1556,
      "mean_token_accuracy": 0.1888101413846016,
      "num_tokens": 66413713.0,
      "step": 32000
    },
    {
      "entropy": 5.560084104537964,
      "epoch": 2.903211175616836,
      "grad_norm": 1.375,
      "learning_rate": 0.0004163177908722262,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18807704746723175,
      "num_tokens": 66424802.0,
      "step": 32005
    },
    {
      "entropy": 5.536942815780639,
      "epoch": 2.90366473149492,
      "grad_norm": 1.4375,
      "learning_rate": 0.00041629261127841105,
      "loss": 5.1694,
      "mean_token_accuracy": 0.19450151175260544,
      "num_tokens": 66434716.0,
      "step": 32010
    },
    {
      "entropy": 5.557676839828491,
      "epoch": 2.9041182873730045,
      "grad_norm": 1.25,
      "learning_rate": 0.00041626742876262876,
      "loss": 5.2433,
      "mean_token_accuracy": 0.18838555365800858,
      "num_tokens": 66445725.0,
      "step": 32015
    },
    {
      "entropy": 5.657399129867554,
      "epoch": 2.9045718432510883,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041624224332539997,
      "loss": 5.2833,
      "mean_token_accuracy": 0.17893050462007523,
      "num_tokens": 66457492.0,
      "step": 32020
    },
    {
      "entropy": 5.682863330841064,
      "epoch": 2.9050253991291726,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004162170549672455,
      "loss": 5.2178,
      "mean_token_accuracy": 0.18817773461341858,
      "num_tokens": 66467490.0,
      "step": 32025
    },
    {
      "entropy": 5.643955421447754,
      "epoch": 2.905478955007257,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004161918636886865,
      "loss": 5.322,
      "mean_token_accuracy": 0.18296095877885818,
      "num_tokens": 66478160.0,
      "step": 32030
    },
    {
      "entropy": 5.612471103668213,
      "epoch": 2.905932510885341,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004161666694902437,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18929250687360763,
      "num_tokens": 66488181.0,
      "step": 32035
    },
    {
      "entropy": 5.576140975952148,
      "epoch": 2.9063860667634254,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004161414723724383,
      "loss": 5.2998,
      "mean_token_accuracy": 0.1886634945869446,
      "num_tokens": 66498507.0,
      "step": 32040
    },
    {
      "entropy": 5.507584142684936,
      "epoch": 2.9068396226415096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004161162723357912,
      "loss": 5.1714,
      "mean_token_accuracy": 0.19025631099939347,
      "num_tokens": 66507752.0,
      "step": 32045
    },
    {
      "entropy": 5.593355178833008,
      "epoch": 2.9072931785195935,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004160910693808236,
      "loss": 5.2431,
      "mean_token_accuracy": 0.1832992687821388,
      "num_tokens": 66519360.0,
      "step": 32050
    },
    {
      "entropy": 5.705951881408692,
      "epoch": 2.9077467343976777,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00041606586350805686,
      "loss": 5.3081,
      "mean_token_accuracy": 0.18640866726636887,
      "num_tokens": 66529790.0,
      "step": 32055
    },
    {
      "entropy": 5.543341016769409,
      "epoch": 2.908200290275762,
      "grad_norm": 1.625,
      "learning_rate": 0.000416040654718012,
      "loss": 5.1629,
      "mean_token_accuracy": 0.20065186470746993,
      "num_tokens": 66539617.0,
      "step": 32060
    },
    {
      "entropy": 5.570460462570191,
      "epoch": 2.9086538461538463,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041601544301121045,
      "loss": 5.1883,
      "mean_token_accuracy": 0.19179969429969787,
      "num_tokens": 66549861.0,
      "step": 32065
    },
    {
      "entropy": 5.6819864273071286,
      "epoch": 2.90910740203193,
      "grad_norm": 2.0,
      "learning_rate": 0.0004159902283881737,
      "loss": 5.3671,
      "mean_token_accuracy": 0.1833731099963188,
      "num_tokens": 66560388.0,
      "step": 32070
    },
    {
      "entropy": 5.65114459991455,
      "epoch": 2.9095609579100143,
      "grad_norm": 1.53125,
      "learning_rate": 0.000415965010849423,
      "loss": 5.3229,
      "mean_token_accuracy": 0.17732912302017212,
      "num_tokens": 66570795.0,
      "step": 32075
    },
    {
      "entropy": 5.666976833343506,
      "epoch": 2.9100145137880986,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041593979039548,
      "loss": 5.2562,
      "mean_token_accuracy": 0.1865604892373085,
      "num_tokens": 66580268.0,
      "step": 32080
    },
    {
      "entropy": 5.656290292739868,
      "epoch": 2.910468069666183,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004159145670268662,
      "loss": 5.3222,
      "mean_token_accuracy": 0.18246627748012542,
      "num_tokens": 66591192.0,
      "step": 32085
    },
    {
      "entropy": 5.619631004333496,
      "epoch": 2.910921625544267,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004158893407441033,
      "loss": 5.3096,
      "mean_token_accuracy": 0.19548963904380798,
      "num_tokens": 66600768.0,
      "step": 32090
    },
    {
      "entropy": 5.679549932479858,
      "epoch": 2.9113751814223514,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00041586411154771285,
      "loss": 5.3176,
      "mean_token_accuracy": 0.18336862325668335,
      "num_tokens": 66612207.0,
      "step": 32095
    },
    {
      "entropy": 5.532453012466431,
      "epoch": 2.9118287373004357,
      "grad_norm": 1.71875,
      "learning_rate": 0.00041583887943821667,
      "loss": 5.1828,
      "mean_token_accuracy": 0.19073133766651154,
      "num_tokens": 66622175.0,
      "step": 32100
    },
    {
      "entropy": 5.591123247146607,
      "epoch": 2.9122822931785195,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004158136444161367,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18850943446159363,
      "num_tokens": 66633936.0,
      "step": 32105
    },
    {
      "entropy": 5.685386848449707,
      "epoch": 2.9127358490566038,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041578840648199455,
      "loss": 5.2712,
      "mean_token_accuracy": 0.18008900731801986,
      "num_tokens": 66643502.0,
      "step": 32110
    },
    {
      "entropy": 5.586417722702026,
      "epoch": 2.913189404934688,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004157631656363124,
      "loss": 5.2557,
      "mean_token_accuracy": 0.1876880943775177,
      "num_tokens": 66654890.0,
      "step": 32115
    },
    {
      "entropy": 5.619087600708008,
      "epoch": 2.9136429608127723,
      "grad_norm": 1.515625,
      "learning_rate": 0.000415737921879612,
      "loss": 5.3279,
      "mean_token_accuracy": 0.173168908059597,
      "num_tokens": 66667037.0,
      "step": 32120
    },
    {
      "entropy": 5.611325645446778,
      "epoch": 2.914096516690856,
      "grad_norm": 1.75,
      "learning_rate": 0.00041571267521241554,
      "loss": 5.2214,
      "mean_token_accuracy": 0.1860846608877182,
      "num_tokens": 66677291.0,
      "step": 32125
    },
    {
      "entropy": 5.524058675765991,
      "epoch": 2.9145500725689404,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004156874256352451,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18669843077659606,
      "num_tokens": 66687421.0,
      "step": 32130
    },
    {
      "entropy": 5.692731237411499,
      "epoch": 2.9150036284470247,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004156621731486228,
      "loss": 5.2957,
      "mean_token_accuracy": 0.18464346826076508,
      "num_tokens": 66697190.0,
      "step": 32135
    },
    {
      "entropy": 5.621494722366333,
      "epoch": 2.915457184325109,
      "grad_norm": 1.3125,
      "learning_rate": 0.000415636917753071,
      "loss": 5.2951,
      "mean_token_accuracy": 0.18687107115983964,
      "num_tokens": 66707255.0,
      "step": 32140
    },
    {
      "entropy": 5.532164001464844,
      "epoch": 2.915910740203193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004156116594491119,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19184435456991195,
      "num_tokens": 66716451.0,
      "step": 32145
    },
    {
      "entropy": 5.611624479293823,
      "epoch": 2.9163642960812775,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004155863982372678,
      "loss": 5.2708,
      "mean_token_accuracy": 0.19079528450965882,
      "num_tokens": 66727157.0,
      "step": 32150
    },
    {
      "entropy": 5.682557773590088,
      "epoch": 2.9168178519593613,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041556113411806117,
      "loss": 5.2578,
      "mean_token_accuracy": 0.1877954751253128,
      "num_tokens": 66737523.0,
      "step": 32155
    },
    {
      "entropy": 5.603590106964111,
      "epoch": 2.9172714078374455,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00041553586709201453,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18859922736883164,
      "num_tokens": 66748183.0,
      "step": 32160
    },
    {
      "entropy": 5.630386018753052,
      "epoch": 2.91772496371553,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004155105971596503,
      "loss": 5.2083,
      "mean_token_accuracy": 0.18983011394739152,
      "num_tokens": 66758794.0,
      "step": 32165
    },
    {
      "entropy": 5.6657885074615475,
      "epoch": 2.918178519593614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041548532432149105,
      "loss": 5.375,
      "mean_token_accuracy": 0.18810697346925737,
      "num_tokens": 66769916.0,
      "step": 32170
    },
    {
      "entropy": 5.601858282089234,
      "epoch": 2.918632075471698,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004154600485780595,
      "loss": 5.2637,
      "mean_token_accuracy": 0.19202176630496978,
      "num_tokens": 66779658.0,
      "step": 32175
    },
    {
      "entropy": 5.5499461650848385,
      "epoch": 2.919085631349782,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041543476992987843,
      "loss": 5.1887,
      "mean_token_accuracy": 0.19235433638095856,
      "num_tokens": 66788909.0,
      "step": 32180
    },
    {
      "entropy": 5.678054475784302,
      "epoch": 2.9195391872278664,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041540948837747056,
      "loss": 5.3004,
      "mean_token_accuracy": 0.18901946842670442,
      "num_tokens": 66799519.0,
      "step": 32185
    },
    {
      "entropy": 5.52849178314209,
      "epoch": 2.9199927431059507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004153842039213586,
      "loss": 5.164,
      "mean_token_accuracy": 0.1899101883172989,
      "num_tokens": 66808915.0,
      "step": 32190
    },
    {
      "entropy": 5.638445091247559,
      "epoch": 2.920446298984035,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004153589165620656,
      "loss": 5.281,
      "mean_token_accuracy": 0.18452817648649217,
      "num_tokens": 66818819.0,
      "step": 32195
    },
    {
      "entropy": 5.517433452606201,
      "epoch": 2.9208998548621192,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041533362630011435,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2015189751982689,
      "num_tokens": 66829274.0,
      "step": 32200
    },
    {
      "entropy": 5.643275547027588,
      "epoch": 2.921353410740203,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000415308333136028,
      "loss": 5.2915,
      "mean_token_accuracy": 0.18863751888275146,
      "num_tokens": 66839659.0,
      "step": 32205
    },
    {
      "entropy": 5.694063282012939,
      "epoch": 2.9218069666182873,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041528303707032957,
      "loss": 5.3304,
      "mean_token_accuracy": 0.18022509813308715,
      "num_tokens": 66850415.0,
      "step": 32210
    },
    {
      "entropy": 5.613066720962524,
      "epoch": 2.9222605224963716,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004152577381035422,
      "loss": 5.1866,
      "mean_token_accuracy": 0.20050296038389206,
      "num_tokens": 66860663.0,
      "step": 32215
    },
    {
      "entropy": 5.629782199859619,
      "epoch": 2.922714078374456,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041523243623618907,
      "loss": 5.247,
      "mean_token_accuracy": 0.1885818600654602,
      "num_tokens": 66871473.0,
      "step": 32220
    },
    {
      "entropy": 5.540227699279785,
      "epoch": 2.9231676342525397,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041520713146879333,
      "loss": 5.2014,
      "mean_token_accuracy": 0.19078682065010072,
      "num_tokens": 66881609.0,
      "step": 32225
    },
    {
      "entropy": 5.633756875991821,
      "epoch": 2.923621190130624,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041518182380187853,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18594983965158463,
      "num_tokens": 66891780.0,
      "step": 32230
    },
    {
      "entropy": 5.632365036010742,
      "epoch": 2.924074746008708,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00041515651323596774,
      "loss": 5.3134,
      "mean_token_accuracy": 0.183112208545208,
      "num_tokens": 66902251.0,
      "step": 32235
    },
    {
      "entropy": 5.65935640335083,
      "epoch": 2.9245283018867925,
      "grad_norm": 1.71875,
      "learning_rate": 0.00041513119977158464,
      "loss": 5.2851,
      "mean_token_accuracy": 0.1850234314799309,
      "num_tokens": 66912538.0,
      "step": 32240
    },
    {
      "entropy": 5.6093751907348635,
      "epoch": 2.9249818577648767,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004151058834092526,
      "loss": 5.202,
      "mean_token_accuracy": 0.19217139184474946,
      "num_tokens": 66922648.0,
      "step": 32245
    },
    {
      "entropy": 5.568642425537109,
      "epoch": 2.925435413642961,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00041508056414949515,
      "loss": 5.1725,
      "mean_token_accuracy": 0.1887093663215637,
      "num_tokens": 66933115.0,
      "step": 32250
    },
    {
      "entropy": 5.682587051391602,
      "epoch": 2.9258889695210453,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000415055241992836,
      "loss": 5.373,
      "mean_token_accuracy": 0.18156035393476486,
      "num_tokens": 66943727.0,
      "step": 32255
    },
    {
      "entropy": 5.593780660629273,
      "epoch": 2.926342525399129,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004150299169397987,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1890658676624298,
      "num_tokens": 66953488.0,
      "step": 32260
    },
    {
      "entropy": 5.630441045761108,
      "epoch": 2.9267960812772134,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004150045889909071,
      "loss": 5.3487,
      "mean_token_accuracy": 0.17777331322431564,
      "num_tokens": 66964586.0,
      "step": 32265
    },
    {
      "entropy": 5.663705348968506,
      "epoch": 2.9272496371552976,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004149792581466848,
      "loss": 5.2616,
      "mean_token_accuracy": 0.18271004259586335,
      "num_tokens": 66975032.0,
      "step": 32270
    },
    {
      "entropy": 5.6170941352844235,
      "epoch": 2.9277031930333814,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041495392440765586,
      "loss": 5.214,
      "mean_token_accuracy": 0.18997307866811752,
      "num_tokens": 66984910.0,
      "step": 32275
    },
    {
      "entropy": 5.525366735458374,
      "epoch": 2.9281567489114657,
      "grad_norm": 2.40625,
      "learning_rate": 0.00041492858777434407,
      "loss": 5.0613,
      "mean_token_accuracy": 0.197720904648304,
      "num_tokens": 66995238.0,
      "step": 32280
    },
    {
      "entropy": 5.555583715438843,
      "epoch": 2.92861030478955,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004149032482472734,
      "loss": 5.2086,
      "mean_token_accuracy": 0.19997346997261048,
      "num_tokens": 67005596.0,
      "step": 32285
    },
    {
      "entropy": 5.6765669822692875,
      "epoch": 2.9290638606676342,
      "grad_norm": 2.875,
      "learning_rate": 0.00041487790582696803,
      "loss": 5.4148,
      "mean_token_accuracy": 0.17206655144691468,
      "num_tokens": 67017312.0,
      "step": 32290
    },
    {
      "entropy": 5.636444711685181,
      "epoch": 2.9295174165457185,
      "grad_norm": 1.25,
      "learning_rate": 0.0004148525605139518,
      "loss": 5.199,
      "mean_token_accuracy": 0.19461503475904465,
      "num_tokens": 67026692.0,
      "step": 32295
    },
    {
      "entropy": 5.531598234176636,
      "epoch": 2.9299709724238028,
      "grad_norm": 1.375,
      "learning_rate": 0.00041482721230874897,
      "loss": 5.2403,
      "mean_token_accuracy": 0.19093101620674133,
      "num_tokens": 67036801.0,
      "step": 32300
    },
    {
      "entropy": 5.525400924682617,
      "epoch": 2.930424528301887,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041480186121188383,
      "loss": 5.1895,
      "mean_token_accuracy": 0.19238993972539903,
      "num_tokens": 67047262.0,
      "step": 32305
    },
    {
      "entropy": 5.662450551986694,
      "epoch": 2.930878084179971,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004147765072238806,
      "loss": 5.2989,
      "mean_token_accuracy": 0.18848079890012742,
      "num_tokens": 67058019.0,
      "step": 32310
    },
    {
      "entropy": 5.659448051452637,
      "epoch": 2.931331640058055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004147511503452635,
      "loss": 5.3201,
      "mean_token_accuracy": 0.1830838829278946,
      "num_tokens": 67068844.0,
      "step": 32315
    },
    {
      "entropy": 5.588466835021973,
      "epoch": 2.9317851959361394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041472579057655703,
      "loss": 5.1741,
      "mean_token_accuracy": 0.195682854950428,
      "num_tokens": 67077400.0,
      "step": 32320
    },
    {
      "entropy": 5.590085411071778,
      "epoch": 2.9322387518142237,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004147004279182856,
      "loss": 5.3799,
      "mean_token_accuracy": 0.1745907485485077,
      "num_tokens": 67088352.0,
      "step": 32325
    },
    {
      "entropy": 5.678563499450684,
      "epoch": 2.9326923076923075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004146750623709737,
      "loss": 5.3022,
      "mean_token_accuracy": 0.18362828046083451,
      "num_tokens": 67097879.0,
      "step": 32330
    },
    {
      "entropy": 5.667727851867676,
      "epoch": 2.9331458635703918,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041464969393514595,
      "loss": 5.1993,
      "mean_token_accuracy": 0.19282942414283752,
      "num_tokens": 67108355.0,
      "step": 32335
    },
    {
      "entropy": 5.564042091369629,
      "epoch": 2.933599419448476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004146243226113269,
      "loss": 5.2852,
      "mean_token_accuracy": 0.18336547017097474,
      "num_tokens": 67118348.0,
      "step": 32340
    },
    {
      "entropy": 5.50234808921814,
      "epoch": 2.9340529753265603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041459894840004135,
      "loss": 5.1596,
      "mean_token_accuracy": 0.19605503231287003,
      "num_tokens": 67128944.0,
      "step": 32345
    },
    {
      "entropy": 5.632363557815552,
      "epoch": 2.9345065312046446,
      "grad_norm": 1.515625,
      "learning_rate": 0.00041457357130181396,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1862534761428833,
      "num_tokens": 67139365.0,
      "step": 32350
    },
    {
      "entropy": 5.620897960662842,
      "epoch": 2.934960087082729,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00041454819131716945,
      "loss": 5.2344,
      "mean_token_accuracy": 0.1836485743522644,
      "num_tokens": 67150406.0,
      "step": 32355
    },
    {
      "entropy": 5.679247093200684,
      "epoch": 2.9354136429608126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041452280844663295,
      "loss": 5.4142,
      "mean_token_accuracy": 0.17848606258630753,
      "num_tokens": 67161631.0,
      "step": 32360
    },
    {
      "entropy": 5.532937908172608,
      "epoch": 2.935867198838897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004144974226907291,
      "loss": 5.249,
      "mean_token_accuracy": 0.1897963359951973,
      "num_tokens": 67171735.0,
      "step": 32365
    },
    {
      "entropy": 5.625122737884522,
      "epoch": 2.936320754716981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000414472034049983,
      "loss": 5.2848,
      "mean_token_accuracy": 0.18658307045698166,
      "num_tokens": 67181616.0,
      "step": 32370
    },
    {
      "entropy": 5.651636552810669,
      "epoch": 2.9367743105950654,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004144466425249197,
      "loss": 5.2049,
      "mean_token_accuracy": 0.19137027412652968,
      "num_tokens": 67192519.0,
      "step": 32375
    },
    {
      "entropy": 5.571226119995117,
      "epoch": 2.9372278664731493,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004144212481160643,
      "loss": 5.2303,
      "mean_token_accuracy": 0.186860491335392,
      "num_tokens": 67202805.0,
      "step": 32380
    },
    {
      "entropy": 5.621926927566529,
      "epoch": 2.9376814223512335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004143958508239419,
      "loss": 5.2247,
      "mean_token_accuracy": 0.18662799745798112,
      "num_tokens": 67212778.0,
      "step": 32385
    },
    {
      "entropy": 5.550153398513794,
      "epoch": 2.938134978229318,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004143704506490779,
      "loss": 5.3033,
      "mean_token_accuracy": 0.18890799283981324,
      "num_tokens": 67222588.0,
      "step": 32390
    },
    {
      "entropy": 5.614031744003296,
      "epoch": 2.938588534107402,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004143450475919975,
      "loss": 5.2745,
      "mean_token_accuracy": 0.18358280807733535,
      "num_tokens": 67233022.0,
      "step": 32395
    },
    {
      "entropy": 5.614336252212524,
      "epoch": 2.9390420899854863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004143196416532259,
      "loss": 5.2147,
      "mean_token_accuracy": 0.19752923548221588,
      "num_tokens": 67242631.0,
      "step": 32400
    },
    {
      "entropy": 5.585832643508911,
      "epoch": 2.9394956458635706,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004142942328332887,
      "loss": 5.1889,
      "mean_token_accuracy": 0.1967334896326065,
      "num_tokens": 67252376.0,
      "step": 32405
    },
    {
      "entropy": 5.558678483963012,
      "epoch": 2.9399492017416544,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004142688211327111,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18471298068761827,
      "num_tokens": 67262209.0,
      "step": 32410
    },
    {
      "entropy": 5.644162845611572,
      "epoch": 2.9404027576197387,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004142434065520189,
      "loss": 5.2518,
      "mean_token_accuracy": 0.18282662332057953,
      "num_tokens": 67272893.0,
      "step": 32415
    },
    {
      "entropy": 5.593526124954224,
      "epoch": 2.940856313497823,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004142179890917376,
      "loss": 5.2169,
      "mean_token_accuracy": 0.18484945893287658,
      "num_tokens": 67284017.0,
      "step": 32420
    },
    {
      "entropy": 5.538451528549194,
      "epoch": 2.941309869375907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041419256875239275,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1943913847208023,
      "num_tokens": 67293674.0,
      "step": 32425
    },
    {
      "entropy": 5.599263763427734,
      "epoch": 2.941763425253991,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004141671455345101,
      "loss": 5.2622,
      "mean_token_accuracy": 0.18935995250940324,
      "num_tokens": 67303759.0,
      "step": 32430
    },
    {
      "entropy": 5.685354852676392,
      "epoch": 2.9422169811320753,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041414171943861543,
      "loss": 5.2823,
      "mean_token_accuracy": 0.19217592626810073,
      "num_tokens": 67314962.0,
      "step": 32435
    },
    {
      "entropy": 5.637516021728516,
      "epoch": 2.9426705370101596,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004141162904652346,
      "loss": 5.3459,
      "mean_token_accuracy": 0.1803679347038269,
      "num_tokens": 67325210.0,
      "step": 32440
    },
    {
      "entropy": 5.639649820327759,
      "epoch": 2.943124092888244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004140908586148934,
      "loss": 5.2684,
      "mean_token_accuracy": 0.1848199963569641,
      "num_tokens": 67335110.0,
      "step": 32445
    },
    {
      "entropy": 5.618850564956665,
      "epoch": 2.943577648766328,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004140654238881178,
      "loss": 5.3556,
      "mean_token_accuracy": 0.18447889983654023,
      "num_tokens": 67345872.0,
      "step": 32450
    },
    {
      "entropy": 5.623377275466919,
      "epoch": 2.9440312046444124,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041403998628543375,
      "loss": 5.2259,
      "mean_token_accuracy": 0.18388163596391677,
      "num_tokens": 67356532.0,
      "step": 32455
    },
    {
      "entropy": 5.568531560897827,
      "epoch": 2.9444847605224966,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041401454580736744,
      "loss": 5.2529,
      "mean_token_accuracy": 0.19116266965866088,
      "num_tokens": 67366603.0,
      "step": 32460
    },
    {
      "entropy": 5.634243631362915,
      "epoch": 2.9449383164005805,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004139891024544448,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1842173755168915,
      "num_tokens": 67377487.0,
      "step": 32465
    },
    {
      "entropy": 5.6595619201660154,
      "epoch": 2.9453918722786647,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004139636562271921,
      "loss": 5.3276,
      "mean_token_accuracy": 0.1874126061797142,
      "num_tokens": 67387704.0,
      "step": 32470
    },
    {
      "entropy": 5.659085130691528,
      "epoch": 2.945845428156749,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041393820712613567,
      "loss": 5.2897,
      "mean_token_accuracy": 0.18541717827320098,
      "num_tokens": 67398409.0,
      "step": 32475
    },
    {
      "entropy": 5.635950994491577,
      "epoch": 2.9462989840348333,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041391275515180166,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18791110068559647,
      "num_tokens": 67408905.0,
      "step": 32480
    },
    {
      "entropy": 5.5829918384552,
      "epoch": 2.946752539912917,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004138873003047165,
      "loss": 5.1889,
      "mean_token_accuracy": 0.19769928902387618,
      "num_tokens": 67418466.0,
      "step": 32485
    },
    {
      "entropy": 5.497446203231812,
      "epoch": 2.9472060957910013,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004138618425854066,
      "loss": 5.2144,
      "mean_token_accuracy": 0.1889340788125992,
      "num_tokens": 67428597.0,
      "step": 32490
    },
    {
      "entropy": 5.647193050384521,
      "epoch": 2.9476596516690856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041383638199439824,
      "loss": 5.2488,
      "mean_token_accuracy": 0.18896857351064683,
      "num_tokens": 67439158.0,
      "step": 32495
    },
    {
      "entropy": 5.6827398300170895,
      "epoch": 2.94811320754717,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041381091853221835,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18320436775684357,
      "num_tokens": 67449540.0,
      "step": 32500
    },
    {
      "entropy": 5.578593826293945,
      "epoch": 2.948566763425254,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004137854521993932,
      "loss": 5.1895,
      "mean_token_accuracy": 0.19361255317926407,
      "num_tokens": 67459385.0,
      "step": 32505
    },
    {
      "entropy": 5.576331090927124,
      "epoch": 2.9490203193033384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041375998299644946,
      "loss": 5.2842,
      "mean_token_accuracy": 0.1879651814699173,
      "num_tokens": 67469597.0,
      "step": 32510
    },
    {
      "entropy": 5.52771372795105,
      "epoch": 2.9494738751814222,
      "grad_norm": 1.328125,
      "learning_rate": 0.000413734510923914,
      "loss": 5.178,
      "mean_token_accuracy": 0.18347822725772858,
      "num_tokens": 67479271.0,
      "step": 32515
    },
    {
      "entropy": 5.544347953796387,
      "epoch": 2.9499274310595065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004137090359823135,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19219921231269838,
      "num_tokens": 67488997.0,
      "step": 32520
    },
    {
      "entropy": 5.6275828838348385,
      "epoch": 2.9503809869375908,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041368355817217474,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18965599238872527,
      "num_tokens": 67499507.0,
      "step": 32525
    },
    {
      "entropy": 5.574159574508667,
      "epoch": 2.950834542815675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041365807749402467,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1909639924764633,
      "num_tokens": 67509916.0,
      "step": 32530
    },
    {
      "entropy": 5.554263496398926,
      "epoch": 2.951288098693759,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004136325939483902,
      "loss": 5.1966,
      "mean_token_accuracy": 0.183832611143589,
      "num_tokens": 67519759.0,
      "step": 32535
    },
    {
      "entropy": 5.574406719207763,
      "epoch": 2.951741654571843,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004136071075357984,
      "loss": 5.2238,
      "mean_token_accuracy": 0.19402537494897842,
      "num_tokens": 67529261.0,
      "step": 32540
    },
    {
      "entropy": 5.567534446716309,
      "epoch": 2.9521952104499274,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004135816182567763,
      "loss": 5.1904,
      "mean_token_accuracy": 0.18990111500024795,
      "num_tokens": 67539084.0,
      "step": 32545
    },
    {
      "entropy": 5.594976854324341,
      "epoch": 2.9526487663280117,
      "grad_norm": 1.234375,
      "learning_rate": 0.000413556126111851,
      "loss": 5.2813,
      "mean_token_accuracy": 0.1898380324244499,
      "num_tokens": 67548551.0,
      "step": 32550
    },
    {
      "entropy": 5.641780662536621,
      "epoch": 2.953102322206096,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004135306311015497,
      "loss": 5.2264,
      "mean_token_accuracy": 0.1871751070022583,
      "num_tokens": 67558364.0,
      "step": 32555
    },
    {
      "entropy": 5.626357126235962,
      "epoch": 2.95355587808418,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004135051332263996,
      "loss": 5.2429,
      "mean_token_accuracy": 0.19232044517993926,
      "num_tokens": 67569257.0,
      "step": 32560
    },
    {
      "entropy": 5.560070896148682,
      "epoch": 2.954009433962264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041347963248692814,
      "loss": 5.296,
      "mean_token_accuracy": 0.18189420700073242,
      "num_tokens": 67580135.0,
      "step": 32565
    },
    {
      "entropy": 5.629810619354248,
      "epoch": 2.9544629898403483,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004134541288836626,
      "loss": 5.2079,
      "mean_token_accuracy": 0.1932898595929146,
      "num_tokens": 67589705.0,
      "step": 32570
    },
    {
      "entropy": 5.60604043006897,
      "epoch": 2.9549165457184325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041342862241713034,
      "loss": 5.1796,
      "mean_token_accuracy": 0.1859543055295944,
      "num_tokens": 67599648.0,
      "step": 32575
    },
    {
      "entropy": 5.576606225967407,
      "epoch": 2.955370101596517,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041340311308785886,
      "loss": 5.2027,
      "mean_token_accuracy": 0.18784306347370147,
      "num_tokens": 67609624.0,
      "step": 32580
    },
    {
      "entropy": 5.531143283843994,
      "epoch": 2.9558236574746006,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041337760089637577,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18908244967460633,
      "num_tokens": 67619596.0,
      "step": 32585
    },
    {
      "entropy": 5.701256990432739,
      "epoch": 2.956277213352685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041335208584320865,
      "loss": 5.2551,
      "mean_token_accuracy": 0.1864472359418869,
      "num_tokens": 67629667.0,
      "step": 32590
    },
    {
      "entropy": 5.656514739990234,
      "epoch": 2.956730769230769,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004133265679288851,
      "loss": 5.2423,
      "mean_token_accuracy": 0.1878908470273018,
      "num_tokens": 67640048.0,
      "step": 32595
    },
    {
      "entropy": 5.612515783309936,
      "epoch": 2.9571843251088534,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004133010471539329,
      "loss": 5.2983,
      "mean_token_accuracy": 0.18386557847261428,
      "num_tokens": 67650731.0,
      "step": 32600
    },
    {
      "entropy": 5.598229598999024,
      "epoch": 2.9576378809869377,
      "grad_norm": 1.25,
      "learning_rate": 0.00041327552351887983,
      "loss": 5.2298,
      "mean_token_accuracy": 0.18651547133922577,
      "num_tokens": 67661696.0,
      "step": 32605
    },
    {
      "entropy": 5.619508409500122,
      "epoch": 2.958091436865022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041324999702425366,
      "loss": 5.2694,
      "mean_token_accuracy": 0.18844599574804305,
      "num_tokens": 67671758.0,
      "step": 32610
    },
    {
      "entropy": 5.591750812530518,
      "epoch": 2.9585449927431062,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004132244676705823,
      "loss": 5.2068,
      "mean_token_accuracy": 0.19027095884084702,
      "num_tokens": 67682488.0,
      "step": 32615
    },
    {
      "entropy": 5.6306455612182615,
      "epoch": 2.95899854862119,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004131989354583938,
      "loss": 5.235,
      "mean_token_accuracy": 0.19272213876247407,
      "num_tokens": 67692945.0,
      "step": 32620
    },
    {
      "entropy": 5.6081358909606935,
      "epoch": 2.9594521044992743,
      "grad_norm": 1.109375,
      "learning_rate": 0.000413173400388216,
      "loss": 5.1437,
      "mean_token_accuracy": 0.19206918627023697,
      "num_tokens": 67702924.0,
      "step": 32625
    },
    {
      "entropy": 5.565289163589478,
      "epoch": 2.9599056603773586,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004131478624605771,
      "loss": 5.2651,
      "mean_token_accuracy": 0.1839776635169983,
      "num_tokens": 67714284.0,
      "step": 32630
    },
    {
      "entropy": 5.6069926738739015,
      "epoch": 2.9603592162554424,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004131223216760052,
      "loss": 5.309,
      "mean_token_accuracy": 0.18325879275798798,
      "num_tokens": 67725902.0,
      "step": 32635
    },
    {
      "entropy": 5.636891841888428,
      "epoch": 2.9608127721335267,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004130967780350286,
      "loss": 5.2132,
      "mean_token_accuracy": 0.18203046768903733,
      "num_tokens": 67736268.0,
      "step": 32640
    },
    {
      "entropy": 5.682074117660522,
      "epoch": 2.961266328011611,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004130712315381753,
      "loss": 5.2855,
      "mean_token_accuracy": 0.19344197660684587,
      "num_tokens": 67745709.0,
      "step": 32645
    },
    {
      "entropy": 5.6135517120361325,
      "epoch": 2.961719883889695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041304568218597385,
      "loss": 5.238,
      "mean_token_accuracy": 0.18632990270853042,
      "num_tokens": 67756092.0,
      "step": 32650
    },
    {
      "entropy": 5.6357160091400145,
      "epoch": 2.9621734397677795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041302012997895246,
      "loss": 5.2733,
      "mean_token_accuracy": 0.18344347476959227,
      "num_tokens": 67766224.0,
      "step": 32655
    },
    {
      "entropy": 5.710088491439819,
      "epoch": 2.9626269956458637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004129945749176395,
      "loss": 5.3848,
      "mean_token_accuracy": 0.18396553099155427,
      "num_tokens": 67776787.0,
      "step": 32660
    },
    {
      "entropy": 5.593484830856323,
      "epoch": 2.963080551523948,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041296901700256375,
      "loss": 5.2373,
      "mean_token_accuracy": 0.18657793253660201,
      "num_tokens": 67787924.0,
      "step": 32665
    },
    {
      "entropy": 5.582508373260498,
      "epoch": 2.963534107402032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004129434562342535,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18146776258945466,
      "num_tokens": 67798544.0,
      "step": 32670
    },
    {
      "entropy": 5.575259017944336,
      "epoch": 2.963987663280116,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041291789261323734,
      "loss": 5.2349,
      "mean_token_accuracy": 0.18963664919137954,
      "num_tokens": 67809528.0,
      "step": 32675
    },
    {
      "entropy": 5.536106538772583,
      "epoch": 2.9644412191582004,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041289232614004416,
      "loss": 5.1479,
      "mean_token_accuracy": 0.20046737641096116,
      "num_tokens": 67820263.0,
      "step": 32680
    },
    {
      "entropy": 5.68050594329834,
      "epoch": 2.9648947750362846,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004128667568152024,
      "loss": 5.3059,
      "mean_token_accuracy": 0.18514337837696077,
      "num_tokens": 67830687.0,
      "step": 32685
    },
    {
      "entropy": 5.706602287292481,
      "epoch": 2.9653483309143684,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041284118463924107,
      "loss": 5.3407,
      "mean_token_accuracy": 0.18006487041711808,
      "num_tokens": 67841714.0,
      "step": 32690
    },
    {
      "entropy": 5.553023147583008,
      "epoch": 2.9658018867924527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004128156096126889,
      "loss": 5.2216,
      "mean_token_accuracy": 0.18612637519836425,
      "num_tokens": 67851952.0,
      "step": 32695
    },
    {
      "entropy": 5.542511606216431,
      "epoch": 2.966255442670537,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041279003173607477,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19155048877000808,
      "num_tokens": 67861958.0,
      "step": 32700
    },
    {
      "entropy": 5.525273370742798,
      "epoch": 2.9667089985486212,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004127644510099278,
      "loss": 5.132,
      "mean_token_accuracy": 0.19539920389652252,
      "num_tokens": 67871654.0,
      "step": 32705
    },
    {
      "entropy": 5.518263483047486,
      "epoch": 2.9671625544267055,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004127388674347767,
      "loss": 5.2129,
      "mean_token_accuracy": 0.19366692155599594,
      "num_tokens": 67881678.0,
      "step": 32710
    },
    {
      "entropy": 5.57129135131836,
      "epoch": 2.96761611030479,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041271328101115075,
      "loss": 5.2647,
      "mean_token_accuracy": 0.18304726332426072,
      "num_tokens": 67892663.0,
      "step": 32715
    },
    {
      "entropy": 5.747872638702392,
      "epoch": 2.9680696661828736,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041268769173957915,
      "loss": 5.4144,
      "mean_token_accuracy": 0.18063365668058395,
      "num_tokens": 67903308.0,
      "step": 32720
    },
    {
      "entropy": 5.655269765853882,
      "epoch": 2.968523222060958,
      "grad_norm": 1.515625,
      "learning_rate": 0.00041266209962059097,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18821370154619216,
      "num_tokens": 67913092.0,
      "step": 32725
    },
    {
      "entropy": 5.600697755813599,
      "epoch": 2.968976777939042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004126365046547154,
      "loss": 5.2355,
      "mean_token_accuracy": 0.19383729100227357,
      "num_tokens": 67923708.0,
      "step": 32730
    },
    {
      "entropy": 5.528707218170166,
      "epoch": 2.9694303338171264,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004126109068424819,
      "loss": 5.2148,
      "mean_token_accuracy": 0.1913328602910042,
      "num_tokens": 67934164.0,
      "step": 32735
    },
    {
      "entropy": 5.633414936065674,
      "epoch": 2.96988388969521,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004125853061844198,
      "loss": 5.3198,
      "mean_token_accuracy": 0.18427855670452117,
      "num_tokens": 67943162.0,
      "step": 32740
    },
    {
      "entropy": 5.704927825927735,
      "epoch": 2.9703374455732945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004125597026810585,
      "loss": 5.31,
      "mean_token_accuracy": 0.1869204044342041,
      "num_tokens": 67953943.0,
      "step": 32745
    },
    {
      "entropy": 5.535838222503662,
      "epoch": 2.9707910014513788,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041253409633292747,
      "loss": 5.1994,
      "mean_token_accuracy": 0.1931968465447426,
      "num_tokens": 67964603.0,
      "step": 32750
    },
    {
      "entropy": 5.678432464599609,
      "epoch": 2.971244557329463,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041250848714055626,
      "loss": 5.2868,
      "mean_token_accuracy": 0.1878018856048584,
      "num_tokens": 67974541.0,
      "step": 32755
    },
    {
      "entropy": 5.670205402374267,
      "epoch": 2.9716981132075473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004124828751044746,
      "loss": 5.25,
      "mean_token_accuracy": 0.18434316217899321,
      "num_tokens": 67984968.0,
      "step": 32760
    },
    {
      "entropy": 5.526771783828735,
      "epoch": 2.9721516690856316,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004124572602252119,
      "loss": 5.1939,
      "mean_token_accuracy": 0.19241692870855331,
      "num_tokens": 67995567.0,
      "step": 32765
    },
    {
      "entropy": 5.674709177017212,
      "epoch": 2.9726052249637154,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004124316425032981,
      "loss": 5.386,
      "mean_token_accuracy": 0.17242269814014435,
      "num_tokens": 68008530.0,
      "step": 32770
    },
    {
      "entropy": 5.67949333190918,
      "epoch": 2.9730587808417996,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041240602193926294,
      "loss": 5.3129,
      "mean_token_accuracy": 0.18436773270368575,
      "num_tokens": 68018393.0,
      "step": 32775
    },
    {
      "entropy": 5.615158271789551,
      "epoch": 2.973512336719884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004123803985336362,
      "loss": 5.2468,
      "mean_token_accuracy": 0.18558162748813628,
      "num_tokens": 68027677.0,
      "step": 32780
    },
    {
      "entropy": 5.540374851226806,
      "epoch": 2.973965892597968,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00041235477228694786,
      "loss": 5.1375,
      "mean_token_accuracy": 0.1960857853293419,
      "num_tokens": 68037157.0,
      "step": 32785
    },
    {
      "entropy": 5.5981996059417725,
      "epoch": 2.974419448476052,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041232914319972775,
      "loss": 5.2376,
      "mean_token_accuracy": 0.19002051055431365,
      "num_tokens": 68047793.0,
      "step": 32790
    },
    {
      "entropy": 5.612902688980102,
      "epoch": 2.9748730043541363,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041230351127250603,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1903377056121826,
      "num_tokens": 68058155.0,
      "step": 32795
    },
    {
      "entropy": 5.625335788726806,
      "epoch": 2.9753265602322205,
      "grad_norm": 1.703125,
      "learning_rate": 0.00041227787650581266,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1904533714056015,
      "num_tokens": 68067440.0,
      "step": 32800
    },
    {
      "entropy": 5.5734148025512695,
      "epoch": 2.975780116110305,
      "grad_norm": 1.125,
      "learning_rate": 0.00041225223890017785,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1873741030693054,
      "num_tokens": 68077815.0,
      "step": 32805
    },
    {
      "entropy": 5.621134948730469,
      "epoch": 2.976233671988389,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00041222659845613173,
      "loss": 5.1945,
      "mean_token_accuracy": 0.19153795391321182,
      "num_tokens": 68087692.0,
      "step": 32810
    },
    {
      "entropy": 5.556379175186157,
      "epoch": 2.9766872278664733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041220095517420464,
      "loss": 5.1815,
      "mean_token_accuracy": 0.19385472536087037,
      "num_tokens": 68097774.0,
      "step": 32815
    },
    {
      "entropy": 5.578641700744629,
      "epoch": 2.9771407837445576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004121753090549268,
      "loss": 5.278,
      "mean_token_accuracy": 0.1847276955842972,
      "num_tokens": 68107992.0,
      "step": 32820
    },
    {
      "entropy": 5.55187463760376,
      "epoch": 2.9775943396226414,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041214966009882857,
      "loss": 5.1766,
      "mean_token_accuracy": 0.1956209197640419,
      "num_tokens": 68117103.0,
      "step": 32825
    },
    {
      "entropy": 5.6413493156433105,
      "epoch": 2.9780478955007257,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004121240083064405,
      "loss": 5.3051,
      "mean_token_accuracy": 0.1873838409781456,
      "num_tokens": 68127951.0,
      "step": 32830
    },
    {
      "entropy": 5.652062940597534,
      "epoch": 2.97850145137881,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041209835367829293,
      "loss": 5.2919,
      "mean_token_accuracy": 0.19266725778579713,
      "num_tokens": 68138229.0,
      "step": 32835
    },
    {
      "entropy": 5.498995208740235,
      "epoch": 2.9789550072568938,
      "grad_norm": 1.25,
      "learning_rate": 0.00041207269621491654,
      "loss": 5.1956,
      "mean_token_accuracy": 0.19157551527023314,
      "num_tokens": 68148019.0,
      "step": 32840
    },
    {
      "entropy": 5.582193326950073,
      "epoch": 2.979408563134978,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004120470359168418,
      "loss": 5.2267,
      "mean_token_accuracy": 0.1834825947880745,
      "num_tokens": 68158704.0,
      "step": 32845
    },
    {
      "entropy": 5.65025372505188,
      "epoch": 2.9798621190130623,
      "grad_norm": 1.46875,
      "learning_rate": 0.00041202137278459946,
      "loss": 5.2557,
      "mean_token_accuracy": 0.1896923750638962,
      "num_tokens": 68169105.0,
      "step": 32850
    },
    {
      "entropy": 5.553771924972534,
      "epoch": 2.9803156748911466,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004119957068187202,
      "loss": 5.1836,
      "mean_token_accuracy": 0.19203780442476273,
      "num_tokens": 68179778.0,
      "step": 32855
    },
    {
      "entropy": 5.564929103851318,
      "epoch": 2.980769230769231,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004119700380197348,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18804909884929658,
      "num_tokens": 68189405.0,
      "step": 32860
    },
    {
      "entropy": 5.5462359428405765,
      "epoch": 2.981222786647315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041194436638817417,
      "loss": 5.1586,
      "mean_token_accuracy": 0.19995884746313095,
      "num_tokens": 68199351.0,
      "step": 32865
    },
    {
      "entropy": 5.641839933395386,
      "epoch": 2.9816763425253994,
      "grad_norm": 1.5,
      "learning_rate": 0.00041191869192456907,
      "loss": 5.3377,
      "mean_token_accuracy": 0.1838509663939476,
      "num_tokens": 68209904.0,
      "step": 32870
    },
    {
      "entropy": 5.674070596694946,
      "epoch": 2.982129898403483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004118930146294506,
      "loss": 5.3271,
      "mean_token_accuracy": 0.18114236146211624,
      "num_tokens": 68221754.0,
      "step": 32875
    },
    {
      "entropy": 5.612284708023071,
      "epoch": 2.9825834542815675,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004118673345033496,
      "loss": 5.2295,
      "mean_token_accuracy": 0.18511561453342437,
      "num_tokens": 68232328.0,
      "step": 32880
    },
    {
      "entropy": 5.534839296340943,
      "epoch": 2.9830370101596517,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004118416515467974,
      "loss": 5.1629,
      "mean_token_accuracy": 0.18860112726688386,
      "num_tokens": 68242812.0,
      "step": 32885
    },
    {
      "entropy": 5.653565406799316,
      "epoch": 2.983490566037736,
      "grad_norm": 1.375,
      "learning_rate": 0.0004118159657603248,
      "loss": 5.3203,
      "mean_token_accuracy": 0.18673077821731568,
      "num_tokens": 68253888.0,
      "step": 32890
    },
    {
      "entropy": 5.628402519226074,
      "epoch": 2.98394412191582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004117902771444633,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18963307440280913,
      "num_tokens": 68264877.0,
      "step": 32895
    },
    {
      "entropy": 5.690961837768555,
      "epoch": 2.984397677793904,
      "grad_norm": 1.140625,
      "learning_rate": 0.000411764585699744,
      "loss": 5.3363,
      "mean_token_accuracy": 0.17806075066328048,
      "num_tokens": 68275863.0,
      "step": 32900
    },
    {
      "entropy": 5.659288120269776,
      "epoch": 2.9848512336719883,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041173889142669806,
      "loss": 5.2569,
      "mean_token_accuracy": 0.184145550429821,
      "num_tokens": 68286768.0,
      "step": 32905
    },
    {
      "entropy": 5.55313401222229,
      "epoch": 2.9853047895500726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004117131943258572,
      "loss": 5.179,
      "mean_token_accuracy": 0.19471507966518403,
      "num_tokens": 68296990.0,
      "step": 32910
    },
    {
      "entropy": 5.498249435424805,
      "epoch": 2.985758345428157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041168749439775255,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1876769095659256,
      "num_tokens": 68307646.0,
      "step": 32915
    },
    {
      "entropy": 5.641096735000611,
      "epoch": 2.986211901306241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004116617916429156,
      "loss": 5.3272,
      "mean_token_accuracy": 0.17423278838396072,
      "num_tokens": 68319586.0,
      "step": 32920
    },
    {
      "entropy": 5.694876718521118,
      "epoch": 2.986665457184325,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004116360860618781,
      "loss": 5.2945,
      "mean_token_accuracy": 0.18898044228553773,
      "num_tokens": 68329692.0,
      "step": 32925
    },
    {
      "entropy": 5.649652290344238,
      "epoch": 2.9871190130624092,
      "grad_norm": 1.75,
      "learning_rate": 0.0004116103776551715,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1859814554452896,
      "num_tokens": 68339589.0,
      "step": 32930
    },
    {
      "entropy": 5.604410839080811,
      "epoch": 2.9875725689404935,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041158466642332743,
      "loss": 5.2662,
      "mean_token_accuracy": 0.17999500334262847,
      "num_tokens": 68349485.0,
      "step": 32935
    },
    {
      "entropy": 5.609663963317871,
      "epoch": 2.9880261248185778,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004115589523668777,
      "loss": 5.2022,
      "mean_token_accuracy": 0.19510260224342346,
      "num_tokens": 68359695.0,
      "step": 32940
    },
    {
      "entropy": 5.564793062210083,
      "epoch": 2.9884796806966616,
      "grad_norm": 1.203125,
      "learning_rate": 0.000411533235486354,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19850803762674332,
      "num_tokens": 68369591.0,
      "step": 32945
    },
    {
      "entropy": 5.638455772399903,
      "epoch": 2.988933236574746,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004115075157822883,
      "loss": 5.2566,
      "mean_token_accuracy": 0.19033820927143097,
      "num_tokens": 68380382.0,
      "step": 32950
    },
    {
      "entropy": 5.471579599380493,
      "epoch": 2.98938679245283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004114817932552122,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19355875998735428,
      "num_tokens": 68389865.0,
      "step": 32955
    },
    {
      "entropy": 5.560301494598389,
      "epoch": 2.9898403483309144,
      "grad_norm": 1.25,
      "learning_rate": 0.00041145606790565797,
      "loss": 5.1727,
      "mean_token_accuracy": 0.1986171841621399,
      "num_tokens": 68399267.0,
      "step": 32960
    },
    {
      "entropy": 5.645993661880493,
      "epoch": 2.9902939042089987,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004114303397341575,
      "loss": 5.2679,
      "mean_token_accuracy": 0.19352024644613267,
      "num_tokens": 68409480.0,
      "step": 32965
    },
    {
      "entropy": 5.700229549407959,
      "epoch": 2.990747460087083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004114046087412427,
      "loss": 5.3375,
      "mean_token_accuracy": 0.18333526998758315,
      "num_tokens": 68420019.0,
      "step": 32970
    },
    {
      "entropy": 5.532309293746948,
      "epoch": 2.9912010159651667,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000411378874927446,
      "loss": 5.1895,
      "mean_token_accuracy": 0.19446461796760559,
      "num_tokens": 68429967.0,
      "step": 32975
    },
    {
      "entropy": 5.577798652648926,
      "epoch": 2.991654571843251,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004113531382932993,
      "loss": 5.208,
      "mean_token_accuracy": 0.18356614261865617,
      "num_tokens": 68441639.0,
      "step": 32980
    },
    {
      "entropy": 5.641994142532349,
      "epoch": 2.9921081277213353,
      "grad_norm": 1.34375,
      "learning_rate": 0.000411327398839335,
      "loss": 5.2388,
      "mean_token_accuracy": 0.1870266988873482,
      "num_tokens": 68452781.0,
      "step": 32985
    },
    {
      "entropy": 5.570036172866821,
      "epoch": 2.9925616835994195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004113016565660854,
      "loss": 5.1902,
      "mean_token_accuracy": 0.19482046812772752,
      "num_tokens": 68463353.0,
      "step": 32990
    },
    {
      "entropy": 5.525812435150146,
      "epoch": 2.9930152394775034,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00041127591147408274,
      "loss": 5.1912,
      "mean_token_accuracy": 0.1862180560827255,
      "num_tokens": 68473507.0,
      "step": 32995
    },
    {
      "entropy": 5.587801933288574,
      "epoch": 2.9934687953555876,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041125016356385945,
      "loss": 5.2153,
      "mean_token_accuracy": 0.1920802190899849,
      "num_tokens": 68483737.0,
      "step": 33000
    },
    {
      "epoch": 2.9934687953555876,
      "eval_entropy": 5.409616090485414,
      "eval_loss": 5.285337924957275,
      "eval_mean_token_accuracy": 0.19523632531286095,
      "eval_num_tokens": 68483737.0,
      "eval_runtime": 26.7621,
      "eval_samples_per_second": 1321.271,
      "eval_steps_per_second": 165.159,
      "step": 33000
    },
    {
      "entropy": 5.629543304443359,
      "epoch": 2.993922351233672,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041122441283594817,
      "loss": 5.2456,
      "mean_token_accuracy": 0.18840995728969573,
      "num_tokens": 68494013.0,
      "step": 33005
    },
    {
      "entropy": 5.559434509277343,
      "epoch": 2.994375907111756,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004111986592908813,
      "loss": 5.2007,
      "mean_token_accuracy": 0.19468654692173004,
      "num_tokens": 68504361.0,
      "step": 33010
    },
    {
      "entropy": 5.5518604755401615,
      "epoch": 2.9948294629898404,
      "grad_norm": 2.203125,
      "learning_rate": 0.00041117290292919134,
      "loss": 5.1695,
      "mean_token_accuracy": 0.19293753206729888,
      "num_tokens": 68514825.0,
      "step": 33015
    },
    {
      "entropy": 5.573344087600708,
      "epoch": 2.9952830188679247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004111471437514111,
      "loss": 5.1757,
      "mean_token_accuracy": 0.18992343097925185,
      "num_tokens": 68524858.0,
      "step": 33020
    },
    {
      "entropy": 5.565552139282227,
      "epoch": 2.995736574746009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004111213817580733,
      "loss": 5.1374,
      "mean_token_accuracy": 0.19995222091674805,
      "num_tokens": 68534481.0,
      "step": 33025
    },
    {
      "entropy": 5.562335014343262,
      "epoch": 2.996190130624093,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004110956169497106,
      "loss": 5.215,
      "mean_token_accuracy": 0.18513901233673097,
      "num_tokens": 68545436.0,
      "step": 33030
    },
    {
      "entropy": 5.574776983261108,
      "epoch": 2.996643686502177,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004110698493268558,
      "loss": 5.2184,
      "mean_token_accuracy": 0.19524132907390596,
      "num_tokens": 68555434.0,
      "step": 33035
    },
    {
      "entropy": 5.582237100601196,
      "epoch": 2.9970972423802613,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004110440788900419,
      "loss": 5.3269,
      "mean_token_accuracy": 0.18568382263183594,
      "num_tokens": 68566295.0,
      "step": 33040
    },
    {
      "entropy": 5.5220866203308105,
      "epoch": 2.9975507982583456,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041101830563980174,
      "loss": 5.1443,
      "mean_token_accuracy": 0.1979670450091362,
      "num_tokens": 68575927.0,
      "step": 33045
    },
    {
      "entropy": 5.663877391815186,
      "epoch": 2.9980043541364294,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041099252957666843,
      "loss": 5.2541,
      "mean_token_accuracy": 0.19663886576890946,
      "num_tokens": 68586254.0,
      "step": 33050
    },
    {
      "entropy": 5.5235755443573,
      "epoch": 2.9984579100145137,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00041096675070117496,
      "loss": 5.2385,
      "mean_token_accuracy": 0.19142424464225768,
      "num_tokens": 68596881.0,
      "step": 33055
    },
    {
      "entropy": 5.488943719863892,
      "epoch": 2.998911465892598,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041094096901385434,
      "loss": 5.1119,
      "mean_token_accuracy": 0.20155316442251206,
      "num_tokens": 68606875.0,
      "step": 33060
    },
    {
      "entropy": 5.650877904891968,
      "epoch": 2.999365021770682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004109151845152399,
      "loss": 5.2628,
      "mean_token_accuracy": 0.18691863864660263,
      "num_tokens": 68617011.0,
      "step": 33065
    },
    {
      "entropy": 5.636035776138305,
      "epoch": 2.9998185776487665,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004108893972058649,
      "loss": 5.2794,
      "mean_token_accuracy": 0.188504858314991,
      "num_tokens": 68627764.0,
      "step": 33070
    },
    {
      "entropy": 5.61772928237915,
      "epoch": 3.0002721335268503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041086360708626236,
      "loss": 5.2176,
      "mean_token_accuracy": 0.19131811559200287,
      "num_tokens": 68637272.0,
      "step": 33075
    },
    {
      "entropy": 5.53045916557312,
      "epoch": 3.0007256894049346,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041083781415696585,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18794280141592026,
      "num_tokens": 68648058.0,
      "step": 33080
    },
    {
      "entropy": 5.519140958786011,
      "epoch": 3.001179245283019,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004108120184185088,
      "loss": 5.1065,
      "mean_token_accuracy": 0.19275668263435364,
      "num_tokens": 68658174.0,
      "step": 33085
    },
    {
      "entropy": 5.623254585266113,
      "epoch": 3.001632801161103,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041078621987142457,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19825504422187806,
      "num_tokens": 68669281.0,
      "step": 33090
    },
    {
      "entropy": 5.524479150772095,
      "epoch": 3.0020863570391874,
      "grad_norm": 1.375,
      "learning_rate": 0.0004107604185162467,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18994625359773637,
      "num_tokens": 68679879.0,
      "step": 33095
    },
    {
      "entropy": 5.486376047134399,
      "epoch": 3.0025399129172716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00041073461435350877,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19505820721387862,
      "num_tokens": 68689852.0,
      "step": 33100
    },
    {
      "entropy": 5.52686710357666,
      "epoch": 3.0029934687953554,
      "grad_norm": 1.515625,
      "learning_rate": 0.00041070880738374447,
      "loss": 5.1555,
      "mean_token_accuracy": 0.18941990286111832,
      "num_tokens": 68700149.0,
      "step": 33105
    },
    {
      "entropy": 5.5610575675964355,
      "epoch": 3.0034470246734397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041068299760748734,
      "loss": 5.1799,
      "mean_token_accuracy": 0.19325367361307144,
      "num_tokens": 68710837.0,
      "step": 33110
    },
    {
      "entropy": 5.613614320755005,
      "epoch": 3.003900580551524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041065718502527134,
      "loss": 5.2004,
      "mean_token_accuracy": 0.19212881326675416,
      "num_tokens": 68721451.0,
      "step": 33115
    },
    {
      "entropy": 5.546064949035644,
      "epoch": 3.0043541364296082,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041063136963763013,
      "loss": 5.1632,
      "mean_token_accuracy": 0.19540501832962037,
      "num_tokens": 68732911.0,
      "step": 33120
    },
    {
      "entropy": 5.535577774047852,
      "epoch": 3.0048076923076925,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004106055514450977,
      "loss": 5.1116,
      "mean_token_accuracy": 0.20045480281114578,
      "num_tokens": 68742650.0,
      "step": 33125
    },
    {
      "entropy": 5.520584440231323,
      "epoch": 3.0052612481857763,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004105797304482079,
      "loss": 5.1424,
      "mean_token_accuracy": 0.19236589819192887,
      "num_tokens": 68753660.0,
      "step": 33130
    },
    {
      "entropy": 5.489102602005005,
      "epoch": 3.0057148040638606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004105539066474947,
      "loss": 5.0595,
      "mean_token_accuracy": 0.1965118944644928,
      "num_tokens": 68763868.0,
      "step": 33135
    },
    {
      "entropy": 5.518217945098877,
      "epoch": 3.006168359941945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041052808004349214,
      "loss": 5.1493,
      "mean_token_accuracy": 0.1935762420296669,
      "num_tokens": 68773911.0,
      "step": 33140
    },
    {
      "entropy": 5.620065498352051,
      "epoch": 3.006621915820029,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004105022506367344,
      "loss": 5.2788,
      "mean_token_accuracy": 0.18314424455165862,
      "num_tokens": 68786423.0,
      "step": 33145
    },
    {
      "entropy": 5.63297700881958,
      "epoch": 3.0070754716981134,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004104764184277555,
      "loss": 5.2424,
      "mean_token_accuracy": 0.18567989766597748,
      "num_tokens": 68797329.0,
      "step": 33150
    },
    {
      "entropy": 5.669068813323975,
      "epoch": 3.0075290275761972,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004104505834170898,
      "loss": 5.2501,
      "mean_token_accuracy": 0.18711768239736556,
      "num_tokens": 68808447.0,
      "step": 33155
    },
    {
      "entropy": 5.611589241027832,
      "epoch": 3.0079825834542815,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00041042474560527153,
      "loss": 5.2042,
      "mean_token_accuracy": 0.1909803718328476,
      "num_tokens": 68819381.0,
      "step": 33160
    },
    {
      "entropy": 5.483608150482178,
      "epoch": 3.0084361393323658,
      "grad_norm": 1.640625,
      "learning_rate": 0.000410398904992835,
      "loss": 5.0982,
      "mean_token_accuracy": 0.20133507549762725,
      "num_tokens": 68829319.0,
      "step": 33165
    },
    {
      "entropy": 5.519392013549805,
      "epoch": 3.00888969521045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041037306158031464,
      "loss": 5.2293,
      "mean_token_accuracy": 0.1945371836423874,
      "num_tokens": 68838640.0,
      "step": 33170
    },
    {
      "entropy": 5.636011600494385,
      "epoch": 3.0093432510885343,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004103472153682448,
      "loss": 5.1498,
      "mean_token_accuracy": 0.19012827277183533,
      "num_tokens": 68847960.0,
      "step": 33175
    },
    {
      "entropy": 5.536379718780518,
      "epoch": 3.009796806966618,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041032136635716016,
      "loss": 5.1628,
      "mean_token_accuracy": 0.19269364327192307,
      "num_tokens": 68858851.0,
      "step": 33180
    },
    {
      "entropy": 5.496761798858643,
      "epoch": 3.0102503628447024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004102955145475951,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1891405001282692,
      "num_tokens": 68869006.0,
      "step": 33185
    },
    {
      "entropy": 5.67933144569397,
      "epoch": 3.0107039187227866,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041026965994008435,
      "loss": 5.2717,
      "mean_token_accuracy": 0.18305541574954987,
      "num_tokens": 68879606.0,
      "step": 33190
    },
    {
      "entropy": 5.556924962997437,
      "epoch": 3.011157474600871,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041024380253516253,
      "loss": 5.1402,
      "mean_token_accuracy": 0.19996216595172883,
      "num_tokens": 68889862.0,
      "step": 33195
    },
    {
      "entropy": 5.639001131057739,
      "epoch": 3.011611030478955,
      "grad_norm": 1.375,
      "learning_rate": 0.00041021794233336444,
      "loss": 5.2529,
      "mean_token_accuracy": 0.19046804904937745,
      "num_tokens": 68900512.0,
      "step": 33200
    },
    {
      "entropy": 5.618736505508423,
      "epoch": 3.012064586357039,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004101920793352249,
      "loss": 5.1873,
      "mean_token_accuracy": 0.1894857957959175,
      "num_tokens": 68910475.0,
      "step": 33205
    },
    {
      "entropy": 5.646018934249878,
      "epoch": 3.0125181422351233,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004101662135412786,
      "loss": 5.2364,
      "mean_token_accuracy": 0.19117489606142044,
      "num_tokens": 68920583.0,
      "step": 33210
    },
    {
      "entropy": 5.514548635482788,
      "epoch": 3.0129716981132075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004101403449520606,
      "loss": 5.1083,
      "mean_token_accuracy": 0.1916455939412117,
      "num_tokens": 68931016.0,
      "step": 33215
    },
    {
      "entropy": 5.536703205108642,
      "epoch": 3.013425253991292,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041011447356810584,
      "loss": 5.1456,
      "mean_token_accuracy": 0.19261294603347778,
      "num_tokens": 68941489.0,
      "step": 33220
    },
    {
      "entropy": 5.60815463066101,
      "epoch": 3.013878809869376,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004100885993899493,
      "loss": 5.255,
      "mean_token_accuracy": 0.18909602761268615,
      "num_tokens": 68952322.0,
      "step": 33225
    },
    {
      "entropy": 5.597724533081054,
      "epoch": 3.01433236574746,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004100627224181261,
      "loss": 5.2268,
      "mean_token_accuracy": 0.19128953963518142,
      "num_tokens": 68962570.0,
      "step": 33230
    },
    {
      "entropy": 5.604410457611084,
      "epoch": 3.014785921625544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041003684265317143,
      "loss": 5.1688,
      "mean_token_accuracy": 0.1922922044992447,
      "num_tokens": 68973733.0,
      "step": 33235
    },
    {
      "entropy": 5.561827993392944,
      "epoch": 3.0152394775036284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041001096009562045,
      "loss": 5.1725,
      "mean_token_accuracy": 0.1994192987680435,
      "num_tokens": 68983526.0,
      "step": 33240
    },
    {
      "entropy": 5.627265787124633,
      "epoch": 3.0156930333817127,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004099850747460084,
      "loss": 5.2623,
      "mean_token_accuracy": 0.19449805617332458,
      "num_tokens": 68993968.0,
      "step": 33245
    },
    {
      "entropy": 5.569505310058593,
      "epoch": 3.016146589259797,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00040995918660487055,
      "loss": 5.1476,
      "mean_token_accuracy": 0.20090068131685257,
      "num_tokens": 69003599.0,
      "step": 33250
    },
    {
      "entropy": 5.540292501449585,
      "epoch": 3.0166001451378808,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004099332956727424,
      "loss": 5.1455,
      "mean_token_accuracy": 0.19184768050909043,
      "num_tokens": 69013686.0,
      "step": 33255
    },
    {
      "entropy": 5.5311003684997555,
      "epoch": 3.017053701015965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004099074019501593,
      "loss": 5.1658,
      "mean_token_accuracy": 0.19199012815952302,
      "num_tokens": 69023803.0,
      "step": 33260
    },
    {
      "entropy": 5.508577299118042,
      "epoch": 3.0175072568940493,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004098815054376567,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19886449128389358,
      "num_tokens": 69033885.0,
      "step": 33265
    },
    {
      "entropy": 5.589107942581177,
      "epoch": 3.0179608127721336,
      "grad_norm": 1.5,
      "learning_rate": 0.0004098556061357702,
      "loss": 5.1545,
      "mean_token_accuracy": 0.19771653413772583,
      "num_tokens": 69044471.0,
      "step": 33270
    },
    {
      "entropy": 5.560167407989502,
      "epoch": 3.018414368650218,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040982970404503544,
      "loss": 5.0451,
      "mean_token_accuracy": 0.1997107595205307,
      "num_tokens": 69055109.0,
      "step": 33275
    },
    {
      "entropy": 5.494177913665771,
      "epoch": 3.018867924528302,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040980379916598803,
      "loss": 5.1617,
      "mean_token_accuracy": 0.19182003140449524,
      "num_tokens": 69064692.0,
      "step": 33280
    },
    {
      "entropy": 5.577710580825806,
      "epoch": 3.019321480406386,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004097778914991637,
      "loss": 5.1703,
      "mean_token_accuracy": 0.1926643431186676,
      "num_tokens": 69075331.0,
      "step": 33285
    },
    {
      "entropy": 5.546709108352661,
      "epoch": 3.01977503628447,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040975198104509827,
      "loss": 5.1561,
      "mean_token_accuracy": 0.19308862388134002,
      "num_tokens": 69085017.0,
      "step": 33290
    },
    {
      "entropy": 5.527767705917358,
      "epoch": 3.0202285921625545,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004097260678043275,
      "loss": 5.1584,
      "mean_token_accuracy": 0.1897374287247658,
      "num_tokens": 69096128.0,
      "step": 33295
    },
    {
      "entropy": 5.6259886741638185,
      "epoch": 3.0206821480406387,
      "grad_norm": 1.453125,
      "learning_rate": 0.00040970015177738725,
      "loss": 5.2416,
      "mean_token_accuracy": 0.18980184942483902,
      "num_tokens": 69106479.0,
      "step": 33300
    },
    {
      "entropy": 5.678152275085449,
      "epoch": 3.021135703918723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004096742329648136,
      "loss": 5.2188,
      "mean_token_accuracy": 0.18724646121263505,
      "num_tokens": 69116346.0,
      "step": 33305
    },
    {
      "entropy": 5.550343561172485,
      "epoch": 3.021589259796807,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004096483113671425,
      "loss": 5.222,
      "mean_token_accuracy": 0.18952141106128692,
      "num_tokens": 69126489.0,
      "step": 33310
    },
    {
      "entropy": 5.558371877670288,
      "epoch": 3.022042815674891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040962238698490994,
      "loss": 5.1574,
      "mean_token_accuracy": 0.19660432785749435,
      "num_tokens": 69136618.0,
      "step": 33315
    },
    {
      "entropy": 5.634967994689942,
      "epoch": 3.0224963715529753,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004095964598186522,
      "loss": 5.2113,
      "mean_token_accuracy": 0.1877411812543869,
      "num_tokens": 69146242.0,
      "step": 33320
    },
    {
      "entropy": 5.59296064376831,
      "epoch": 3.0229499274310596,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004095705298689053,
      "loss": 5.1681,
      "mean_token_accuracy": 0.19200789034366608,
      "num_tokens": 69157269.0,
      "step": 33325
    },
    {
      "entropy": 5.645405006408692,
      "epoch": 3.023403483309144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040954459713620563,
      "loss": 5.2633,
      "mean_token_accuracy": 0.1852481856942177,
      "num_tokens": 69169260.0,
      "step": 33330
    },
    {
      "entropy": 5.643528938293457,
      "epoch": 3.0238570391872277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004095186616210893,
      "loss": 5.2848,
      "mean_token_accuracy": 0.17905704379081727,
      "num_tokens": 69179815.0,
      "step": 33335
    },
    {
      "entropy": 5.628597354888916,
      "epoch": 3.024310595065312,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004094927233240928,
      "loss": 5.2587,
      "mean_token_accuracy": 0.1916957288980484,
      "num_tokens": 69189959.0,
      "step": 33340
    },
    {
      "entropy": 5.571043014526367,
      "epoch": 3.0247641509433962,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040946678224575264,
      "loss": 5.1424,
      "mean_token_accuracy": 0.19467613399028777,
      "num_tokens": 69199485.0,
      "step": 33345
    },
    {
      "entropy": 5.51661901473999,
      "epoch": 3.0252177068214805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000409440838386605,
      "loss": 5.1668,
      "mean_token_accuracy": 0.19478238224983216,
      "num_tokens": 69209180.0,
      "step": 33350
    },
    {
      "entropy": 5.586281490325928,
      "epoch": 3.0256712626995648,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004094148917471866,
      "loss": 5.2267,
      "mean_token_accuracy": 0.18581755757331847,
      "num_tokens": 69220648.0,
      "step": 33355
    },
    {
      "entropy": 5.521419191360474,
      "epoch": 3.0261248185776486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000409388942328034,
      "loss": 5.0386,
      "mean_token_accuracy": 0.19648581892251968,
      "num_tokens": 69230685.0,
      "step": 33360
    },
    {
      "entropy": 5.506122684478759,
      "epoch": 3.026578374455733,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004093629901296838,
      "loss": 5.2121,
      "mean_token_accuracy": 0.1872626483440399,
      "num_tokens": 69240407.0,
      "step": 33365
    },
    {
      "entropy": 5.704458951950073,
      "epoch": 3.027031930333817,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004093370351526727,
      "loss": 5.2451,
      "mean_token_accuracy": 0.18581774085760117,
      "num_tokens": 69250624.0,
      "step": 33370
    },
    {
      "entropy": 5.619555854797364,
      "epoch": 3.0274854862119014,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004093110773975376,
      "loss": 5.2462,
      "mean_token_accuracy": 0.18202880769968033,
      "num_tokens": 69261131.0,
      "step": 33375
    },
    {
      "entropy": 5.498091697692871,
      "epoch": 3.0279390420899857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004092851168648152,
      "loss": 5.2041,
      "mean_token_accuracy": 0.18374566733837128,
      "num_tokens": 69270668.0,
      "step": 33380
    },
    {
      "entropy": 5.568058347702026,
      "epoch": 3.0283925979680695,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004092591535550422,
      "loss": 5.1134,
      "mean_token_accuracy": 0.2027987316250801,
      "num_tokens": 69280528.0,
      "step": 33385
    },
    {
      "entropy": 5.543421220779419,
      "epoch": 3.0288461538461537,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004092331874687558,
      "loss": 5.1503,
      "mean_token_accuracy": 0.19359390288591385,
      "num_tokens": 69290376.0,
      "step": 33390
    },
    {
      "entropy": 5.577026271820069,
      "epoch": 3.029299709724238,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040920721860649285,
      "loss": 5.1817,
      "mean_token_accuracy": 0.2025150865316391,
      "num_tokens": 69301766.0,
      "step": 33395
    },
    {
      "entropy": 5.619555234909058,
      "epoch": 3.0297532656023223,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004091812469687904,
      "loss": 5.1475,
      "mean_token_accuracy": 0.18969434648752212,
      "num_tokens": 69312668.0,
      "step": 33400
    },
    {
      "entropy": 5.617461824417115,
      "epoch": 3.0302068214804065,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004091552725561856,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18754076063632966,
      "num_tokens": 69323195.0,
      "step": 33405
    },
    {
      "entropy": 5.541171455383301,
      "epoch": 3.0306603773584904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004091292953692156,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19969502687454224,
      "num_tokens": 69333812.0,
      "step": 33410
    },
    {
      "entropy": 5.558510732650757,
      "epoch": 3.0311139332365746,
      "grad_norm": 1.578125,
      "learning_rate": 0.00040910331540841755,
      "loss": 5.183,
      "mean_token_accuracy": 0.20181493014097213,
      "num_tokens": 69344378.0,
      "step": 33415
    },
    {
      "entropy": 5.536408233642578,
      "epoch": 3.031567489114659,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00040907733267432877,
      "loss": 5.1773,
      "mean_token_accuracy": 0.19089210033416748,
      "num_tokens": 69354592.0,
      "step": 33420
    },
    {
      "entropy": 5.59657130241394,
      "epoch": 3.032021044992743,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00040905134716748656,
      "loss": 5.1748,
      "mean_token_accuracy": 0.18918184638023378,
      "num_tokens": 69365011.0,
      "step": 33425
    },
    {
      "entropy": 5.522092485427857,
      "epoch": 3.0324746008708274,
      "grad_norm": 1.75,
      "learning_rate": 0.0004090253588884284,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1870290607213974,
      "num_tokens": 69377010.0,
      "step": 33430
    },
    {
      "entropy": 5.580898332595825,
      "epoch": 3.0329281567489113,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040899936783769157,
      "loss": 5.1897,
      "mean_token_accuracy": 0.19270147681236266,
      "num_tokens": 69387263.0,
      "step": 33435
    },
    {
      "entropy": 5.575372695922852,
      "epoch": 3.0333817126269955,
      "grad_norm": 1.578125,
      "learning_rate": 0.00040897337401581374,
      "loss": 5.231,
      "mean_token_accuracy": 0.1904039740562439,
      "num_tokens": 69399188.0,
      "step": 33440
    },
    {
      "entropy": 5.53751859664917,
      "epoch": 3.03383526850508,
      "grad_norm": 1.125,
      "learning_rate": 0.00040894737742333244,
      "loss": 5.1748,
      "mean_token_accuracy": 0.18826807141304017,
      "num_tokens": 69408271.0,
      "step": 33445
    },
    {
      "entropy": 5.635159349441528,
      "epoch": 3.034288824383164,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004089213780607851,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18297690153121948,
      "num_tokens": 69418424.0,
      "step": 33450
    },
    {
      "entropy": 5.602784252166748,
      "epoch": 3.0347423802612483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004088953759287096,
      "loss": 5.1746,
      "mean_token_accuracy": 0.19787155389785765,
      "num_tokens": 69427802.0,
      "step": 33455
    },
    {
      "entropy": 5.547683191299439,
      "epoch": 3.0351959361393326,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004088693710276436,
      "loss": 5.2196,
      "mean_token_accuracy": 0.19588661044836045,
      "num_tokens": 69437581.0,
      "step": 33460
    },
    {
      "entropy": 5.56308331489563,
      "epoch": 3.0356494920174164,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040884336335812494,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1994319885969162,
      "num_tokens": 69447711.0,
      "step": 33465
    },
    {
      "entropy": 5.61976203918457,
      "epoch": 3.0361030478955007,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004088173529206914,
      "loss": 5.1192,
      "mean_token_accuracy": 0.19540292918682098,
      "num_tokens": 69457692.0,
      "step": 33470
    },
    {
      "entropy": 5.518042469024659,
      "epoch": 3.036556603773585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004087913397158809,
      "loss": 5.133,
      "mean_token_accuracy": 0.1915278196334839,
      "num_tokens": 69468034.0,
      "step": 33475
    },
    {
      "entropy": 5.532874011993409,
      "epoch": 3.037010159651669,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040876532374423135,
      "loss": 5.1315,
      "mean_token_accuracy": 0.19902484714984894,
      "num_tokens": 69477719.0,
      "step": 33480
    },
    {
      "entropy": 5.599833869934082,
      "epoch": 3.0374637155297535,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004087393050062809,
      "loss": 5.1675,
      "mean_token_accuracy": 0.18785958141088485,
      "num_tokens": 69487261.0,
      "step": 33485
    },
    {
      "entropy": 5.580426454544067,
      "epoch": 3.0379172714078373,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040871328350256745,
      "loss": 5.1406,
      "mean_token_accuracy": 0.19118794351816176,
      "num_tokens": 69496658.0,
      "step": 33490
    },
    {
      "entropy": 5.630440425872803,
      "epoch": 3.0383708272859216,
      "grad_norm": 1.46875,
      "learning_rate": 0.00040868725923362927,
      "loss": 5.3197,
      "mean_token_accuracy": 0.18411619067192078,
      "num_tokens": 69507197.0,
      "step": 33495
    },
    {
      "entropy": 5.55195574760437,
      "epoch": 3.038824383164006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040866123220000455,
      "loss": 5.2094,
      "mean_token_accuracy": 0.19668736606836318,
      "num_tokens": 69516905.0,
      "step": 33500
    },
    {
      "entropy": 5.60170841217041,
      "epoch": 3.03927793904209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040863520240223144,
      "loss": 5.1273,
      "mean_token_accuracy": 0.19815970212221146,
      "num_tokens": 69526869.0,
      "step": 33505
    },
    {
      "entropy": 5.567000675201416,
      "epoch": 3.0397314949201744,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004086091698408483,
      "loss": 5.1855,
      "mean_token_accuracy": 0.1927609086036682,
      "num_tokens": 69537058.0,
      "step": 33510
    },
    {
      "entropy": 5.636953258514405,
      "epoch": 3.040185050798258,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040858313451639343,
      "loss": 5.2556,
      "mean_token_accuracy": 0.18353887796401977,
      "num_tokens": 69547617.0,
      "step": 33515
    },
    {
      "entropy": 5.644690895080567,
      "epoch": 3.0406386066763424,
      "grad_norm": 1.5,
      "learning_rate": 0.0004085570964294053,
      "loss": 5.2371,
      "mean_token_accuracy": 0.1813416063785553,
      "num_tokens": 69558733.0,
      "step": 33520
    },
    {
      "entropy": 5.682887411117553,
      "epoch": 3.0410921625544267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004085310555804225,
      "loss": 5.2801,
      "mean_token_accuracy": 0.1856962502002716,
      "num_tokens": 69570011.0,
      "step": 33525
    },
    {
      "entropy": 5.645456027984619,
      "epoch": 3.041545718432511,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00040850501196998336,
      "loss": 5.323,
      "mean_token_accuracy": 0.1794607475399971,
      "num_tokens": 69580894.0,
      "step": 33530
    },
    {
      "entropy": 5.6164466381073,
      "epoch": 3.0419992743105952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040847896559862654,
      "loss": 5.2132,
      "mean_token_accuracy": 0.18779074847698213,
      "num_tokens": 69591438.0,
      "step": 33535
    },
    {
      "entropy": 5.598085403442383,
      "epoch": 3.042452830188679,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004084529164668907,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1899029344320297,
      "num_tokens": 69601745.0,
      "step": 33540
    },
    {
      "entropy": 5.572098302841186,
      "epoch": 3.0429063860667633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040842686457531453,
      "loss": 5.1617,
      "mean_token_accuracy": 0.19726267009973525,
      "num_tokens": 69611872.0,
      "step": 33545
    },
    {
      "entropy": 5.569802284240723,
      "epoch": 3.0433599419448476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004084008099244368,
      "loss": 5.154,
      "mean_token_accuracy": 0.19044438302516936,
      "num_tokens": 69621673.0,
      "step": 33550
    },
    {
      "entropy": 5.529368925094604,
      "epoch": 3.043813497822932,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040837475251479633,
      "loss": 5.1096,
      "mean_token_accuracy": 0.19583346843719482,
      "num_tokens": 69632090.0,
      "step": 33555
    },
    {
      "entropy": 5.538460206985474,
      "epoch": 3.044267053701016,
      "grad_norm": 1.28125,
      "learning_rate": 0.000408348692346932,
      "loss": 5.1417,
      "mean_token_accuracy": 0.19660040140151977,
      "num_tokens": 69642920.0,
      "step": 33560
    },
    {
      "entropy": 5.61982421875,
      "epoch": 3.0447206095791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00040832262942138277,
      "loss": 5.274,
      "mean_token_accuracy": 0.1891783818602562,
      "num_tokens": 69653936.0,
      "step": 33565
    },
    {
      "entropy": 5.5532020092010494,
      "epoch": 3.0451741654571842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004082965637386875,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19819516390562059,
      "num_tokens": 69663362.0,
      "step": 33570
    },
    {
      "entropy": 5.544570207595825,
      "epoch": 3.0456277213352685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004082704952993854,
      "loss": 5.1268,
      "mean_token_accuracy": 0.1956744074821472,
      "num_tokens": 69673178.0,
      "step": 33575
    },
    {
      "entropy": 5.631007432937622,
      "epoch": 3.0460812772133528,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004082444241040155,
      "loss": 5.2223,
      "mean_token_accuracy": 0.19002043455839157,
      "num_tokens": 69683627.0,
      "step": 33580
    },
    {
      "entropy": 5.573653697967529,
      "epoch": 3.046534833091437,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004082183501531169,
      "loss": 5.1854,
      "mean_token_accuracy": 0.193665212392807,
      "num_tokens": 69694010.0,
      "step": 33585
    },
    {
      "entropy": 5.557163095474243,
      "epoch": 3.046988388969521,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00040819227344722893,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18559281527996063,
      "num_tokens": 69704888.0,
      "step": 33590
    },
    {
      "entropy": 5.5317377090454105,
      "epoch": 3.047441944847605,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040816619398689074,
      "loss": 5.1692,
      "mean_token_accuracy": 0.19718896150588988,
      "num_tokens": 69714718.0,
      "step": 33595
    },
    {
      "entropy": 5.590826606750488,
      "epoch": 3.0478955007256894,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040814011177264174,
      "loss": 5.1557,
      "mean_token_accuracy": 0.19602789133787155,
      "num_tokens": 69726076.0,
      "step": 33600
    },
    {
      "entropy": 5.546645641326904,
      "epoch": 3.0483490566037736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040811402680502137,
      "loss": 5.127,
      "mean_token_accuracy": 0.20021650344133377,
      "num_tokens": 69736507.0,
      "step": 33605
    },
    {
      "entropy": 5.492884683609009,
      "epoch": 3.048802612481858,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040808793908456885,
      "loss": 5.1464,
      "mean_token_accuracy": 0.19283103793859482,
      "num_tokens": 69746061.0,
      "step": 33610
    },
    {
      "entropy": 5.5690656185150145,
      "epoch": 3.0492561683599417,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004080618486118241,
      "loss": 5.2207,
      "mean_token_accuracy": 0.18880419582128524,
      "num_tokens": 69756493.0,
      "step": 33615
    },
    {
      "entropy": 5.613162612915039,
      "epoch": 3.049709724238026,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004080357553873261,
      "loss": 5.1273,
      "mean_token_accuracy": 0.1943557605147362,
      "num_tokens": 69766937.0,
      "step": 33620
    },
    {
      "entropy": 5.548803424835205,
      "epoch": 3.0501632801161103,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000408009659411615,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19244129210710526,
      "num_tokens": 69776872.0,
      "step": 33625
    },
    {
      "entropy": 5.584054088592529,
      "epoch": 3.0506168359941945,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004079835606852301,
      "loss": 5.1694,
      "mean_token_accuracy": 0.19157444685697556,
      "num_tokens": 69788145.0,
      "step": 33630
    },
    {
      "entropy": 5.558744430541992,
      "epoch": 3.051070391872279,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004079574592087114,
      "loss": 5.141,
      "mean_token_accuracy": 0.18867748230695724,
      "num_tokens": 69797353.0,
      "step": 33635
    },
    {
      "entropy": 5.543464279174804,
      "epoch": 3.051523947750363,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040793135498259854,
      "loss": 5.1714,
      "mean_token_accuracy": 0.19410113394260406,
      "num_tokens": 69806915.0,
      "step": 33640
    },
    {
      "entropy": 5.573034477233887,
      "epoch": 3.051977503628447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040790524800743134,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1932433143258095,
      "num_tokens": 69817695.0,
      "step": 33645
    },
    {
      "entropy": 5.4945722103118895,
      "epoch": 3.052431059506531,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004078791382837498,
      "loss": 5.0953,
      "mean_token_accuracy": 0.20663119554519654,
      "num_tokens": 69827451.0,
      "step": 33650
    },
    {
      "entropy": 5.607445144653321,
      "epoch": 3.0528846153846154,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040785302581209374,
      "loss": 5.1849,
      "mean_token_accuracy": 0.19585053324699403,
      "num_tokens": 69837897.0,
      "step": 33655
    },
    {
      "entropy": 5.539115333557129,
      "epoch": 3.0533381712626997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040782691059300326,
      "loss": 5.1449,
      "mean_token_accuracy": 0.1912333220243454,
      "num_tokens": 69847378.0,
      "step": 33660
    },
    {
      "entropy": 5.535183477401733,
      "epoch": 3.053791727140784,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004078007926270185,
      "loss": 5.2274,
      "mean_token_accuracy": 0.19128317385911942,
      "num_tokens": 69857522.0,
      "step": 33665
    },
    {
      "entropy": 5.615515184402466,
      "epoch": 3.0542452830188678,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040777467191467946,
      "loss": 5.174,
      "mean_token_accuracy": 0.1927144259214401,
      "num_tokens": 69868384.0,
      "step": 33670
    },
    {
      "entropy": 5.609692764282227,
      "epoch": 3.054698838896952,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004077485484565264,
      "loss": 5.1711,
      "mean_token_accuracy": 0.1955135554075241,
      "num_tokens": 69878761.0,
      "step": 33675
    },
    {
      "entropy": 5.577006053924561,
      "epoch": 3.0551523947750363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004077224222530994,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18922869712114335,
      "num_tokens": 69888366.0,
      "step": 33680
    },
    {
      "entropy": 5.5319007396697994,
      "epoch": 3.0556059506531206,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000407696293304939,
      "loss": 5.1628,
      "mean_token_accuracy": 0.18792270123958588,
      "num_tokens": 69898489.0,
      "step": 33685
    },
    {
      "entropy": 5.5409502506256105,
      "epoch": 3.056059506531205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004076701616125854,
      "loss": 5.1238,
      "mean_token_accuracy": 0.19160011410713196,
      "num_tokens": 69908425.0,
      "step": 33690
    },
    {
      "entropy": 5.575048398971558,
      "epoch": 3.0565130624092887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004076440271765791,
      "loss": 5.2017,
      "mean_token_accuracy": 0.1959786370396614,
      "num_tokens": 69918488.0,
      "step": 33695
    },
    {
      "entropy": 5.516828393936157,
      "epoch": 3.056966618287373,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004076178899974605,
      "loss": 5.0844,
      "mean_token_accuracy": 0.1987137958407402,
      "num_tokens": 69928831.0,
      "step": 33700
    },
    {
      "entropy": 5.62200026512146,
      "epoch": 3.057420174165457,
      "grad_norm": 1.53125,
      "learning_rate": 0.00040759175007577017,
      "loss": 5.1926,
      "mean_token_accuracy": 0.18790875673294066,
      "num_tokens": 69938974.0,
      "step": 33705
    },
    {
      "entropy": 5.510329532623291,
      "epoch": 3.0578737300435415,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004075656074120486,
      "loss": 5.0963,
      "mean_token_accuracy": 0.19339015632867812,
      "num_tokens": 69948804.0,
      "step": 33710
    },
    {
      "entropy": 5.5677885055542,
      "epoch": 3.0583272859216257,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004075394620068365,
      "loss": 5.1698,
      "mean_token_accuracy": 0.18839031904935838,
      "num_tokens": 69958945.0,
      "step": 33715
    },
    {
      "entropy": 5.562092447280884,
      "epoch": 3.0587808417997095,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00040751331386067456,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19754113703966142,
      "num_tokens": 69968826.0,
      "step": 33720
    },
    {
      "entropy": 5.596329689025879,
      "epoch": 3.059234397677794,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004074871629741035,
      "loss": 5.2356,
      "mean_token_accuracy": 0.19170392751693727,
      "num_tokens": 69978961.0,
      "step": 33725
    },
    {
      "entropy": 5.512242174148559,
      "epoch": 3.059687953555878,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004074610093476641,
      "loss": 5.1835,
      "mean_token_accuracy": 0.19285430312156676,
      "num_tokens": 69989848.0,
      "step": 33730
    },
    {
      "entropy": 5.55490779876709,
      "epoch": 3.0601415094339623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040743485298189744,
      "loss": 5.1344,
      "mean_token_accuracy": 0.19320553094148635,
      "num_tokens": 69999607.0,
      "step": 33735
    },
    {
      "entropy": 5.577856492996216,
      "epoch": 3.0605950653120466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040740869387734415,
      "loss": 5.1811,
      "mean_token_accuracy": 0.19240015000104904,
      "num_tokens": 70010446.0,
      "step": 33740
    },
    {
      "entropy": 5.500862789154053,
      "epoch": 3.0610486211901304,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004073825320345453,
      "loss": 5.0992,
      "mean_token_accuracy": 0.18756560534238814,
      "num_tokens": 70021148.0,
      "step": 33745
    },
    {
      "entropy": 5.612251949310303,
      "epoch": 3.0615021770682147,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040735636745404196,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18630332499742508,
      "num_tokens": 70030765.0,
      "step": 33750
    },
    {
      "entropy": 5.665703678131104,
      "epoch": 3.061955732946299,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040733020013637526,
      "loss": 5.1754,
      "mean_token_accuracy": 0.19261312782764434,
      "num_tokens": 70041717.0,
      "step": 33755
    },
    {
      "entropy": 5.602911329269409,
      "epoch": 3.0624092888243832,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00040730403008208633,
      "loss": 5.2454,
      "mean_token_accuracy": 0.18829857856035231,
      "num_tokens": 70052178.0,
      "step": 33760
    },
    {
      "entropy": 5.550325345993042,
      "epoch": 3.0628628447024675,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004072778572917163,
      "loss": 5.1944,
      "mean_token_accuracy": 0.1877523198723793,
      "num_tokens": 70062819.0,
      "step": 33765
    },
    {
      "entropy": 5.51904411315918,
      "epoch": 3.0633164005805513,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.1804,
      "mean_token_accuracy": 0.1933843031525612,
      "num_tokens": 70073966.0,
      "step": 33770
    },
    {
      "entropy": 5.611111736297607,
      "epoch": 3.0637699564586356,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040722550350489825,
      "loss": 5.1557,
      "mean_token_accuracy": 0.19328408241271972,
      "num_tokens": 70083884.0,
      "step": 33775
    },
    {
      "entropy": 5.628449296951294,
      "epoch": 3.06422351233672,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004071993225095327,
      "loss": 5.162,
      "mean_token_accuracy": 0.19160741716623306,
      "num_tokens": 70094910.0,
      "step": 33780
    },
    {
      "entropy": 5.47262716293335,
      "epoch": 3.064677068214804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004071731387802517,
      "loss": 5.024,
      "mean_token_accuracy": 0.20517876297235488,
      "num_tokens": 70104480.0,
      "step": 33785
    },
    {
      "entropy": 5.504285764694214,
      "epoch": 3.0651306240928884,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004071469523175964,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19268052726984025,
      "num_tokens": 70114003.0,
      "step": 33790
    },
    {
      "entropy": 5.566390132904052,
      "epoch": 3.065584179970972,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040712076312210844,
      "loss": 5.2062,
      "mean_token_accuracy": 0.1916944295167923,
      "num_tokens": 70123604.0,
      "step": 33795
    },
    {
      "entropy": 5.597794246673584,
      "epoch": 3.0660377358490565,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00040709457119432945,
      "loss": 5.1901,
      "mean_token_accuracy": 0.19378910213708878,
      "num_tokens": 70133895.0,
      "step": 33800
    },
    {
      "entropy": 5.630712366104126,
      "epoch": 3.0664912917271407,
      "grad_norm": 2.25,
      "learning_rate": 0.00040706837653480113,
      "loss": 5.174,
      "mean_token_accuracy": 0.19685048907995223,
      "num_tokens": 70145063.0,
      "step": 33805
    },
    {
      "entropy": 5.544713735580444,
      "epoch": 3.066944847605225,
      "grad_norm": 1.53125,
      "learning_rate": 0.000407042179144065,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18870967626571655,
      "num_tokens": 70155456.0,
      "step": 33810
    },
    {
      "entropy": 5.440461158752441,
      "epoch": 3.0673984034833093,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040701597902266306,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19984812587499617,
      "num_tokens": 70165427.0,
      "step": 33815
    },
    {
      "entropy": 5.598884248733521,
      "epoch": 3.067851959361393,
      "grad_norm": 1.28125,
      "learning_rate": 0.000406989776171137,
      "loss": 5.2202,
      "mean_token_accuracy": 0.18936819583177567,
      "num_tokens": 70175161.0,
      "step": 33820
    },
    {
      "entropy": 5.555546665191651,
      "epoch": 3.0683055152394774,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040696357059002877,
      "loss": 5.1512,
      "mean_token_accuracy": 0.2017428234219551,
      "num_tokens": 70185026.0,
      "step": 33825
    },
    {
      "entropy": 5.61548547744751,
      "epoch": 3.0687590711175616,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004069373622798802,
      "loss": 5.1838,
      "mean_token_accuracy": 0.20114363431930543,
      "num_tokens": 70195645.0,
      "step": 33830
    },
    {
      "entropy": 5.607908916473389,
      "epoch": 3.069212626995646,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004069111512412334,
      "loss": 5.2743,
      "mean_token_accuracy": 0.18961944729089736,
      "num_tokens": 70206591.0,
      "step": 33835
    },
    {
      "entropy": 5.628829097747802,
      "epoch": 3.06966618287373,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040688493747463036,
      "loss": 5.2019,
      "mean_token_accuracy": 0.1990969032049179,
      "num_tokens": 70216945.0,
      "step": 33840
    },
    {
      "entropy": 5.543893527984619,
      "epoch": 3.0701197387518144,
      "grad_norm": 1.375,
      "learning_rate": 0.0004068587209806132,
      "loss": 5.0794,
      "mean_token_accuracy": 0.20247486233711243,
      "num_tokens": 70226983.0,
      "step": 33845
    },
    {
      "entropy": 5.507295560836792,
      "epoch": 3.0705732946298983,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040683250175972414,
      "loss": 5.1513,
      "mean_token_accuracy": 0.19393004328012467,
      "num_tokens": 70237730.0,
      "step": 33850
    },
    {
      "entropy": 5.561237573623657,
      "epoch": 3.0710268505079825,
      "grad_norm": 1.25,
      "learning_rate": 0.0004068062798125053,
      "loss": 5.223,
      "mean_token_accuracy": 0.18640146851539613,
      "num_tokens": 70248499.0,
      "step": 33855
    },
    {
      "entropy": 5.541390037536621,
      "epoch": 3.071480406386067,
      "grad_norm": 1.4375,
      "learning_rate": 0.000406780055139499,
      "loss": 5.1531,
      "mean_token_accuracy": 0.1932791292667389,
      "num_tokens": 70259294.0,
      "step": 33860
    },
    {
      "entropy": 5.652671957015992,
      "epoch": 3.071933962264151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004067538277412476,
      "loss": 5.2183,
      "mean_token_accuracy": 0.18848797380924226,
      "num_tokens": 70269400.0,
      "step": 33865
    },
    {
      "entropy": 5.641640901565552,
      "epoch": 3.0723875181422353,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004067275976182935,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1951575607061386,
      "num_tokens": 70279380.0,
      "step": 33870
    },
    {
      "entropy": 5.611793661117554,
      "epoch": 3.072841074020319,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00040670136477117906,
      "loss": 5.1881,
      "mean_token_accuracy": 0.19336883425712587,
      "num_tokens": 70289919.0,
      "step": 33875
    },
    {
      "entropy": 5.633363103866577,
      "epoch": 3.0732946298984034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040667512920044684,
      "loss": 5.2905,
      "mean_token_accuracy": 0.19147884100675583,
      "num_tokens": 70300136.0,
      "step": 33880
    },
    {
      "entropy": 5.528422546386719,
      "epoch": 3.0737481857764877,
      "grad_norm": 1.546875,
      "learning_rate": 0.00040664889090663944,
      "loss": 5.0765,
      "mean_token_accuracy": 0.2070856511592865,
      "num_tokens": 70310072.0,
      "step": 33885
    },
    {
      "entropy": 5.551430130004883,
      "epoch": 3.074201741654572,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040662264989029933,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19700147658586503,
      "num_tokens": 70320185.0,
      "step": 33890
    },
    {
      "entropy": 5.535780811309815,
      "epoch": 3.074655297532656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004065964061519694,
      "loss": 5.1613,
      "mean_token_accuracy": 0.19616346359252929,
      "num_tokens": 70330211.0,
      "step": 33895
    },
    {
      "entropy": 5.635112476348877,
      "epoch": 3.07510885341074,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004065701596921922,
      "loss": 5.2142,
      "mean_token_accuracy": 0.18727406859397888,
      "num_tokens": 70341044.0,
      "step": 33900
    },
    {
      "entropy": 5.5738777160644535,
      "epoch": 3.0755624092888243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004065439105115106,
      "loss": 5.2241,
      "mean_token_accuracy": 0.19254755079746247,
      "num_tokens": 70351320.0,
      "step": 33905
    },
    {
      "entropy": 5.510387420654297,
      "epoch": 3.0760159651669086,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040651765861046734,
      "loss": 5.0967,
      "mean_token_accuracy": 0.20132653564214706,
      "num_tokens": 70361214.0,
      "step": 33910
    },
    {
      "entropy": 5.620298957824707,
      "epoch": 3.076469521044993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040649140398960537,
      "loss": 5.2179,
      "mean_token_accuracy": 0.19693188816308976,
      "num_tokens": 70371083.0,
      "step": 33915
    },
    {
      "entropy": 5.648729658126831,
      "epoch": 3.076923076923077,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00040646514664946775,
      "loss": 5.3073,
      "mean_token_accuracy": 0.1861311450600624,
      "num_tokens": 70381987.0,
      "step": 33920
    },
    {
      "entropy": 5.567285966873169,
      "epoch": 3.077376632801161,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004064388865905974,
      "loss": 5.1469,
      "mean_token_accuracy": 0.19398923814296723,
      "num_tokens": 70393462.0,
      "step": 33925
    },
    {
      "entropy": 5.487349605560302,
      "epoch": 3.077830188679245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004064126238135373,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20796983540058137,
      "num_tokens": 70403664.0,
      "step": 33930
    },
    {
      "entropy": 5.510803079605102,
      "epoch": 3.0782837445573294,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004063863583188308,
      "loss": 5.1739,
      "mean_token_accuracy": 0.19066118448972702,
      "num_tokens": 70414588.0,
      "step": 33935
    },
    {
      "entropy": 5.52899055480957,
      "epoch": 3.0787373004354137,
      "grad_norm": 1.25,
      "learning_rate": 0.0004063600901070207,
      "loss": 5.2353,
      "mean_token_accuracy": 0.18749598264694214,
      "num_tokens": 70425572.0,
      "step": 33940
    },
    {
      "entropy": 5.556038475036621,
      "epoch": 3.079190856313498,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040633381917865057,
      "loss": 5.1355,
      "mean_token_accuracy": 0.18951034396886826,
      "num_tokens": 70436140.0,
      "step": 33945
    },
    {
      "entropy": 5.561174631118774,
      "epoch": 3.079644412191582,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004063075455342636,
      "loss": 5.1814,
      "mean_token_accuracy": 0.19434715509414674,
      "num_tokens": 70447117.0,
      "step": 33950
    },
    {
      "entropy": 5.575560855865478,
      "epoch": 3.080097968069666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004062812691744032,
      "loss": 5.1472,
      "mean_token_accuracy": 0.1924455612897873,
      "num_tokens": 70457756.0,
      "step": 33955
    },
    {
      "entropy": 5.5903746604919435,
      "epoch": 3.0805515239477503,
      "grad_norm": 1.125,
      "learning_rate": 0.0004062549900996126,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18945187777280809,
      "num_tokens": 70468671.0,
      "step": 33960
    },
    {
      "entropy": 5.572639989852905,
      "epoch": 3.0810050798258346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004062287083104354,
      "loss": 5.1265,
      "mean_token_accuracy": 0.19683998674154282,
      "num_tokens": 70478212.0,
      "step": 33965
    },
    {
      "entropy": 5.505567169189453,
      "epoch": 3.081458635703919,
      "grad_norm": 1.375,
      "learning_rate": 0.00040620242380741504,
      "loss": 5.147,
      "mean_token_accuracy": 0.19206700026988982,
      "num_tokens": 70489062.0,
      "step": 33970
    },
    {
      "entropy": 5.605609130859375,
      "epoch": 3.0819121915820027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004061761365910951,
      "loss": 5.2372,
      "mean_token_accuracy": 0.19250507652759552,
      "num_tokens": 70499732.0,
      "step": 33975
    },
    {
      "entropy": 5.630833864212036,
      "epoch": 3.082365747460087,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040614984666201927,
      "loss": 5.162,
      "mean_token_accuracy": 0.19039867222309112,
      "num_tokens": 70509777.0,
      "step": 33980
    },
    {
      "entropy": 5.595611953735352,
      "epoch": 3.0828193033381712,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004061235540207312,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1895325466990471,
      "num_tokens": 70520674.0,
      "step": 33985
    },
    {
      "entropy": 5.567700958251953,
      "epoch": 3.0832728592162555,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004060972586677746,
      "loss": 5.1859,
      "mean_token_accuracy": 0.187660376727581,
      "num_tokens": 70531115.0,
      "step": 33990
    },
    {
      "entropy": 5.526043319702149,
      "epoch": 3.0837264150943398,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004060709606036933,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19542192667722702,
      "num_tokens": 70540726.0,
      "step": 33995
    },
    {
      "entropy": 5.595890235900879,
      "epoch": 3.084179970972424,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004060446598290311,
      "loss": 5.1437,
      "mean_token_accuracy": 0.19945759922266007,
      "num_tokens": 70551227.0,
      "step": 34000
    },
    {
      "entropy": 5.519315958023071,
      "epoch": 3.084633526850508,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000406018356344332,
      "loss": 5.1414,
      "mean_token_accuracy": 0.19158499240875243,
      "num_tokens": 70560815.0,
      "step": 34005
    },
    {
      "entropy": 5.59052038192749,
      "epoch": 3.085087082728592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004059920501501398,
      "loss": 5.2335,
      "mean_token_accuracy": 0.19299159497022628,
      "num_tokens": 70570247.0,
      "step": 34010
    },
    {
      "entropy": 5.5910224437713625,
      "epoch": 3.0855406386066764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004059657412469987,
      "loss": 5.2046,
      "mean_token_accuracy": 0.19413191378116607,
      "num_tokens": 70581497.0,
      "step": 34015
    },
    {
      "entropy": 5.531876468658448,
      "epoch": 3.0859941944847606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004059394296354527,
      "loss": 5.1737,
      "mean_token_accuracy": 0.1923488512635231,
      "num_tokens": 70593133.0,
      "step": 34020
    },
    {
      "entropy": 5.549392604827881,
      "epoch": 3.086447750362845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004059131153160459,
      "loss": 5.1953,
      "mean_token_accuracy": 0.19294084757566451,
      "num_tokens": 70603953.0,
      "step": 34025
    },
    {
      "entropy": 5.593243217468261,
      "epoch": 3.0869013062409287,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004058867982893225,
      "loss": 5.1395,
      "mean_token_accuracy": 0.2001527279615402,
      "num_tokens": 70615292.0,
      "step": 34030
    },
    {
      "entropy": 5.620580434799194,
      "epoch": 3.087354862119013,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004058604785558268,
      "loss": 5.3075,
      "mean_token_accuracy": 0.1780841112136841,
      "num_tokens": 70626681.0,
      "step": 34035
    },
    {
      "entropy": 5.60079140663147,
      "epoch": 3.0878084179970973,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040583415611610305,
      "loss": 5.2044,
      "mean_token_accuracy": 0.1898583084344864,
      "num_tokens": 70637339.0,
      "step": 34040
    },
    {
      "entropy": 5.621716260910034,
      "epoch": 3.0882619738751815,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004058078309706956,
      "loss": 5.1648,
      "mean_token_accuracy": 0.19475949108600615,
      "num_tokens": 70647980.0,
      "step": 34045
    },
    {
      "entropy": 5.4953062534332275,
      "epoch": 3.088715529753266,
      "grad_norm": 1.125,
      "learning_rate": 0.0004057815031201488,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2099942684173584,
      "num_tokens": 70657042.0,
      "step": 34050
    },
    {
      "entropy": 5.570208024978638,
      "epoch": 3.0891690856313496,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004057551725650073,
      "loss": 5.2256,
      "mean_token_accuracy": 0.18530482053756714,
      "num_tokens": 70668542.0,
      "step": 34055
    },
    {
      "entropy": 5.604976606369019,
      "epoch": 3.089622641509434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040572883930581544,
      "loss": 5.2297,
      "mean_token_accuracy": 0.19199159145355224,
      "num_tokens": 70679332.0,
      "step": 34060
    },
    {
      "entropy": 5.712366914749145,
      "epoch": 3.090076197387518,
      "grad_norm": 1.125,
      "learning_rate": 0.0004057025033431179,
      "loss": 5.3053,
      "mean_token_accuracy": 0.18807628899812698,
      "num_tokens": 70689897.0,
      "step": 34065
    },
    {
      "entropy": 5.531790113449096,
      "epoch": 3.0905297532656024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040567616467745933,
      "loss": 5.1345,
      "mean_token_accuracy": 0.19569239169359207,
      "num_tokens": 70699965.0,
      "step": 34070
    },
    {
      "entropy": 5.576922512054443,
      "epoch": 3.0909833091436867,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004056498233093843,
      "loss": 5.2572,
      "mean_token_accuracy": 0.1931062638759613,
      "num_tokens": 70710583.0,
      "step": 34075
    },
    {
      "entropy": 5.555461454391479,
      "epoch": 3.0914368650217705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040562347923943766,
      "loss": 5.1382,
      "mean_token_accuracy": 0.19859593212604523,
      "num_tokens": 70721800.0,
      "step": 34080
    },
    {
      "entropy": 5.549019718170166,
      "epoch": 3.0918904208998548,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004055971324681642,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18765602856874466,
      "num_tokens": 70731527.0,
      "step": 34085
    },
    {
      "entropy": 5.6256472110748295,
      "epoch": 3.092343976777939,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004055707829961087,
      "loss": 5.21,
      "mean_token_accuracy": 0.19178076684474946,
      "num_tokens": 70740821.0,
      "step": 34090
    },
    {
      "entropy": 5.5212071418762205,
      "epoch": 3.0927975326560233,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040554443082381627,
      "loss": 5.1311,
      "mean_token_accuracy": 0.19733942151069642,
      "num_tokens": 70750506.0,
      "step": 34095
    },
    {
      "entropy": 5.531621599197388,
      "epoch": 3.0932510885341076,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004055180759518315,
      "loss": 5.1871,
      "mean_token_accuracy": 0.19352924227714538,
      "num_tokens": 70762009.0,
      "step": 34100
    },
    {
      "entropy": 5.583073997497559,
      "epoch": 3.0937046444121914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004054917183806999,
      "loss": 5.1517,
      "mean_token_accuracy": 0.19948724210262297,
      "num_tokens": 70772030.0,
      "step": 34105
    },
    {
      "entropy": 5.577005243301391,
      "epoch": 3.0941582002902757,
      "grad_norm": 1.4375,
      "learning_rate": 0.00040546535811096626,
      "loss": 5.217,
      "mean_token_accuracy": 0.1863364800810814,
      "num_tokens": 70782539.0,
      "step": 34110
    },
    {
      "entropy": 5.663269805908203,
      "epoch": 3.09461175616836,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040543899514317564,
      "loss": 5.3367,
      "mean_token_accuracy": 0.1797260731458664,
      "num_tokens": 70793433.0,
      "step": 34115
    },
    {
      "entropy": 5.62370023727417,
      "epoch": 3.095065312046444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004054126294778735,
      "loss": 5.2182,
      "mean_token_accuracy": 0.19061806797981262,
      "num_tokens": 70802919.0,
      "step": 34120
    },
    {
      "entropy": 5.583073711395263,
      "epoch": 3.0955188679245285,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040538626111560487,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1995431438088417,
      "num_tokens": 70813827.0,
      "step": 34125
    },
    {
      "entropy": 5.629304552078247,
      "epoch": 3.0959724238026123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040535989005691515,
      "loss": 5.2633,
      "mean_token_accuracy": 0.19239185899496078,
      "num_tokens": 70824117.0,
      "step": 34130
    },
    {
      "entropy": 5.516493463516236,
      "epoch": 3.0964259796806965,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040533351630234985,
      "loss": 5.1142,
      "mean_token_accuracy": 0.1974860593676567,
      "num_tokens": 70833387.0,
      "step": 34135
    },
    {
      "entropy": 5.558797883987427,
      "epoch": 3.096879535558781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000405307139852454,
      "loss": 5.1493,
      "mean_token_accuracy": 0.19453014433383942,
      "num_tokens": 70844363.0,
      "step": 34140
    },
    {
      "entropy": 5.596268987655639,
      "epoch": 3.097333091436865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040528076070777335,
      "loss": 5.2164,
      "mean_token_accuracy": 0.19027090817689896,
      "num_tokens": 70854759.0,
      "step": 34145
    },
    {
      "entropy": 5.529345989227295,
      "epoch": 3.0977866473149493,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004052543788688534,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1949127048254013,
      "num_tokens": 70865682.0,
      "step": 34150
    },
    {
      "entropy": 5.644157457351684,
      "epoch": 3.098240203193033,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004052279943362396,
      "loss": 5.2337,
      "mean_token_accuracy": 0.18644345849752425,
      "num_tokens": 70875449.0,
      "step": 34155
    },
    {
      "entropy": 5.568035078048706,
      "epoch": 3.0986937590711174,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004052016071104779,
      "loss": 5.17,
      "mean_token_accuracy": 0.19225307255983354,
      "num_tokens": 70885885.0,
      "step": 34160
    },
    {
      "entropy": 5.546226978302002,
      "epoch": 3.0991473149492017,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004051752171921136,
      "loss": 5.1497,
      "mean_token_accuracy": 0.19069391936063768,
      "num_tokens": 70896320.0,
      "step": 34165
    },
    {
      "entropy": 5.566938495635986,
      "epoch": 3.099600870827286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040514882458169267,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19093994945287704,
      "num_tokens": 70907074.0,
      "step": 34170
    },
    {
      "entropy": 5.534225130081177,
      "epoch": 3.1000544267053702,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051224292797609,
      "loss": 5.157,
      "mean_token_accuracy": 0.18771844059228898,
      "num_tokens": 70916567.0,
      "step": 34175
    },
    {
      "entropy": 5.522459268569946,
      "epoch": 3.100507982583454,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040509603128686413,
      "loss": 5.1816,
      "mean_token_accuracy": 0.19327054619789125,
      "num_tokens": 70926163.0,
      "step": 34180
    },
    {
      "entropy": 5.610204887390137,
      "epoch": 3.1009615384615383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004050696306035483,
      "loss": 5.1816,
      "mean_token_accuracy": 0.19252336025238037,
      "num_tokens": 70937358.0,
      "step": 34185
    },
    {
      "entropy": 5.554514217376709,
      "epoch": 3.1014150943396226,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040504322723035937,
      "loss": 5.1679,
      "mean_token_accuracy": 0.19143812656402587,
      "num_tokens": 70949305.0,
      "step": 34190
    },
    {
      "entropy": 5.490369701385498,
      "epoch": 3.101868650217707,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040501682116784326,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19878117740154266,
      "num_tokens": 70959146.0,
      "step": 34195
    },
    {
      "entropy": 5.517150831222534,
      "epoch": 3.102322206095791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040499041241654626,
      "loss": 5.184,
      "mean_token_accuracy": 0.19342068433761597,
      "num_tokens": 70968828.0,
      "step": 34200
    },
    {
      "entropy": 5.55975022315979,
      "epoch": 3.1027757619738754,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004049640009770144,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19252739250659942,
      "num_tokens": 70979733.0,
      "step": 34205
    },
    {
      "entropy": 5.607820320129394,
      "epoch": 3.103229317851959,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004049375868497939,
      "loss": 5.2301,
      "mean_token_accuracy": 0.19293882101774215,
      "num_tokens": 70989726.0,
      "step": 34210
    },
    {
      "entropy": 5.6253063678741455,
      "epoch": 3.1036828737300435,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040491117003543087,
      "loss": 5.2271,
      "mean_token_accuracy": 0.19375976473093032,
      "num_tokens": 71000715.0,
      "step": 34215
    },
    {
      "entropy": 5.485222959518433,
      "epoch": 3.1041364296081277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040488475053447186,
      "loss": 5.0034,
      "mean_token_accuracy": 0.2059576004743576,
      "num_tokens": 71011453.0,
      "step": 34220
    },
    {
      "entropy": 5.552039289474488,
      "epoch": 3.104589985486212,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040485832834746306,
      "loss": 5.1698,
      "mean_token_accuracy": 0.19787020683288575,
      "num_tokens": 71021477.0,
      "step": 34225
    },
    {
      "entropy": 5.543666124343872,
      "epoch": 3.1050435413642963,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040483190347495097,
      "loss": 5.1084,
      "mean_token_accuracy": 0.19498308300971984,
      "num_tokens": 71031706.0,
      "step": 34230
    },
    {
      "entropy": 5.62398738861084,
      "epoch": 3.10549709724238,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000404805475917482,
      "loss": 5.162,
      "mean_token_accuracy": 0.19376360476017,
      "num_tokens": 71042882.0,
      "step": 34235
    },
    {
      "entropy": 5.52817440032959,
      "epoch": 3.1059506531204644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040477904567560275,
      "loss": 5.1324,
      "mean_token_accuracy": 0.20054336339235307,
      "num_tokens": 71052804.0,
      "step": 34240
    },
    {
      "entropy": 5.570269823074341,
      "epoch": 3.1064042089985486,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004047526127498597,
      "loss": 5.2516,
      "mean_token_accuracy": 0.1861145332455635,
      "num_tokens": 71064349.0,
      "step": 34245
    },
    {
      "entropy": 5.577790689468384,
      "epoch": 3.106857764876633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040472617714079964,
      "loss": 5.1738,
      "mean_token_accuracy": 0.197662653028965,
      "num_tokens": 71075754.0,
      "step": 34250
    },
    {
      "entropy": 5.564915513992309,
      "epoch": 3.107311320754717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004046997388489691,
      "loss": 5.1037,
      "mean_token_accuracy": 0.1958732634782791,
      "num_tokens": 71085338.0,
      "step": 34255
    },
    {
      "entropy": 5.484837532043457,
      "epoch": 3.107764876632801,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004046732978749149,
      "loss": 5.1284,
      "mean_token_accuracy": 0.19483007192611695,
      "num_tokens": 71097089.0,
      "step": 34260
    },
    {
      "entropy": 5.5067298889160154,
      "epoch": 3.1082184325108853,
      "grad_norm": 1.15625,
      "learning_rate": 0.000404646854219184,
      "loss": 5.0813,
      "mean_token_accuracy": 0.20316479504108428,
      "num_tokens": 71106990.0,
      "step": 34265
    },
    {
      "entropy": 5.505002546310425,
      "epoch": 3.1086719883889695,
      "grad_norm": 1.0625,
      "learning_rate": 0.000404620407882323,
      "loss": 5.1229,
      "mean_token_accuracy": 0.19436001032590866,
      "num_tokens": 71117634.0,
      "step": 34270
    },
    {
      "entropy": 5.4811155796051025,
      "epoch": 3.109125544267054,
      "grad_norm": 1.25,
      "learning_rate": 0.00040459395886487885,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19328183233737944,
      "num_tokens": 71127717.0,
      "step": 34275
    },
    {
      "entropy": 5.5353141784667965,
      "epoch": 3.109579100145138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004045675071673988,
      "loss": 5.1044,
      "mean_token_accuracy": 0.20143332183361054,
      "num_tokens": 71137172.0,
      "step": 34280
    },
    {
      "entropy": 5.595251178741455,
      "epoch": 3.110032656023222,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004045410527904296,
      "loss": 5.2007,
      "mean_token_accuracy": 0.18896124511957169,
      "num_tokens": 71146843.0,
      "step": 34285
    },
    {
      "entropy": 5.484085941314698,
      "epoch": 3.110486211901306,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004045145957345184,
      "loss": 5.0806,
      "mean_token_accuracy": 0.20217190384864808,
      "num_tokens": 71156619.0,
      "step": 34290
    },
    {
      "entropy": 5.561825513839722,
      "epoch": 3.1109397677793904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004044881360002123,
      "loss": 5.1547,
      "mean_token_accuracy": 0.19503698050975798,
      "num_tokens": 71167234.0,
      "step": 34295
    },
    {
      "entropy": 5.522214078903199,
      "epoch": 3.1113933236574747,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004044616735880586,
      "loss": 5.1379,
      "mean_token_accuracy": 0.19650880992412567,
      "num_tokens": 71178049.0,
      "step": 34300
    },
    {
      "entropy": 5.509947919845581,
      "epoch": 3.111846879535559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040443520849860453,
      "loss": 5.1029,
      "mean_token_accuracy": 0.1907911166548729,
      "num_tokens": 71188568.0,
      "step": 34305
    },
    {
      "entropy": 5.62548246383667,
      "epoch": 3.1123004354136428,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00040440874073239733,
      "loss": 5.1704,
      "mean_token_accuracy": 0.1914988487958908,
      "num_tokens": 71199004.0,
      "step": 34310
    },
    {
      "entropy": 5.565518426895141,
      "epoch": 3.112753991291727,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040438227028998436,
      "loss": 5.1544,
      "mean_token_accuracy": 0.19195429533720015,
      "num_tokens": 71209239.0,
      "step": 34315
    },
    {
      "entropy": 5.527668857574463,
      "epoch": 3.1132075471698113,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00040435579717191313,
      "loss": 5.205,
      "mean_token_accuracy": 0.19153906106948854,
      "num_tokens": 71219469.0,
      "step": 34320
    },
    {
      "entropy": 5.550353860855102,
      "epoch": 3.1136611030478956,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040432932137873107,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18887262493371965,
      "num_tokens": 71230670.0,
      "step": 34325
    },
    {
      "entropy": 5.6299115180969235,
      "epoch": 3.11411465892598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004043028429109857,
      "loss": 5.2801,
      "mean_token_accuracy": 0.19012881964445114,
      "num_tokens": 71241512.0,
      "step": 34330
    },
    {
      "entropy": 5.554801750183105,
      "epoch": 3.1145682148040637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004042763617692245,
      "loss": 5.108,
      "mean_token_accuracy": 0.19996727705001832,
      "num_tokens": 71250869.0,
      "step": 34335
    },
    {
      "entropy": 5.511534261703491,
      "epoch": 3.115021770682148,
      "grad_norm": 1.125,
      "learning_rate": 0.00040424987795399523,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19504529237747192,
      "num_tokens": 71260577.0,
      "step": 34340
    },
    {
      "entropy": 5.507219886779785,
      "epoch": 3.115475326560232,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004042233914658455,
      "loss": 5.1947,
      "mean_token_accuracy": 0.1939687505364418,
      "num_tokens": 71270185.0,
      "step": 34345
    },
    {
      "entropy": 5.5392763137817385,
      "epoch": 3.1159288824383164,
      "grad_norm": 1.25,
      "learning_rate": 0.00040419690230532324,
      "loss": 5.1162,
      "mean_token_accuracy": 0.20081414729356767,
      "num_tokens": 71279427.0,
      "step": 34350
    },
    {
      "entropy": 5.577287912368774,
      "epoch": 3.1163824383164007,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040417041047297606,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19577111601829528,
      "num_tokens": 71290857.0,
      "step": 34355
    },
    {
      "entropy": 5.545188093185425,
      "epoch": 3.116835994194485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004041439159693518,
      "loss": 5.2406,
      "mean_token_accuracy": 0.19534343779087066,
      "num_tokens": 71300971.0,
      "step": 34360
    },
    {
      "entropy": 5.593053197860717,
      "epoch": 3.117289550072569,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004041174187949985,
      "loss": 5.2859,
      "mean_token_accuracy": 0.1874040499329567,
      "num_tokens": 71311017.0,
      "step": 34365
    },
    {
      "entropy": 5.587882709503174,
      "epoch": 3.117743105950653,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00040409091895046413,
      "loss": 5.2611,
      "mean_token_accuracy": 0.18666754961013793,
      "num_tokens": 71321546.0,
      "step": 34370
    },
    {
      "entropy": 5.553908157348633,
      "epoch": 3.1181966618287373,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040406441643629656,
      "loss": 5.1764,
      "mean_token_accuracy": 0.18941422998905183,
      "num_tokens": 71332345.0,
      "step": 34375
    },
    {
      "entropy": 5.671159172058106,
      "epoch": 3.1186502177068216,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040403791125304405,
      "loss": 5.3049,
      "mean_token_accuracy": 0.18542334139347078,
      "num_tokens": 71342670.0,
      "step": 34380
    },
    {
      "entropy": 5.597239065170288,
      "epoch": 3.119103773584906,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004040114034012546,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19392962008714676,
      "num_tokens": 71353440.0,
      "step": 34385
    },
    {
      "entropy": 5.54554500579834,
      "epoch": 3.1195573294629897,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040398489288147646,
      "loss": 5.2318,
      "mean_token_accuracy": 0.19338992834091187,
      "num_tokens": 71363199.0,
      "step": 34390
    },
    {
      "entropy": 5.574510383605957,
      "epoch": 3.120010885341074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040395837969425786,
      "loss": 5.1624,
      "mean_token_accuracy": 0.19093506634235383,
      "num_tokens": 71372925.0,
      "step": 34395
    },
    {
      "entropy": 5.608377170562744,
      "epoch": 3.1204644412191582,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004039318638401471,
      "loss": 5.1813,
      "mean_token_accuracy": 0.1973096802830696,
      "num_tokens": 71383350.0,
      "step": 34400
    },
    {
      "entropy": 5.526504993438721,
      "epoch": 3.1209179970972425,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004039053453196926,
      "loss": 5.1255,
      "mean_token_accuracy": 0.19963813573122025,
      "num_tokens": 71394006.0,
      "step": 34405
    },
    {
      "entropy": 5.481306266784668,
      "epoch": 3.1213715529753268,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004038788241334427,
      "loss": 5.1592,
      "mean_token_accuracy": 0.1942255213856697,
      "num_tokens": 71403945.0,
      "step": 34410
    },
    {
      "entropy": 5.565738582611084,
      "epoch": 3.1218251088534106,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004038523002819458,
      "loss": 5.1712,
      "mean_token_accuracy": 0.1866019681096077,
      "num_tokens": 71414238.0,
      "step": 34415
    },
    {
      "entropy": 5.535795640945435,
      "epoch": 3.122278664731495,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040382577376575055,
      "loss": 5.1756,
      "mean_token_accuracy": 0.1876346528530121,
      "num_tokens": 71423586.0,
      "step": 34420
    },
    {
      "entropy": 5.582615375518799,
      "epoch": 3.122732220609579,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004037992445854054,
      "loss": 5.1329,
      "mean_token_accuracy": 0.19640420973300934,
      "num_tokens": 71434261.0,
      "step": 34425
    },
    {
      "entropy": 5.547533130645752,
      "epoch": 3.1231857764876634,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040377271274145916,
      "loss": 5.1479,
      "mean_token_accuracy": 0.19757260829210282,
      "num_tokens": 71443952.0,
      "step": 34430
    },
    {
      "entropy": 5.584706449508667,
      "epoch": 3.1236393323657476,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004037461782344603,
      "loss": 5.1473,
      "mean_token_accuracy": 0.19150156974792482,
      "num_tokens": 71454051.0,
      "step": 34435
    },
    {
      "entropy": 5.5157617092132565,
      "epoch": 3.1240928882438315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040371964106495777,
      "loss": 5.1452,
      "mean_token_accuracy": 0.19569385349750518,
      "num_tokens": 71464579.0,
      "step": 34440
    },
    {
      "entropy": 5.54561071395874,
      "epoch": 3.1245464441219157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040369310123350015,
      "loss": 5.1588,
      "mean_token_accuracy": 0.19428974986076356,
      "num_tokens": 71474631.0,
      "step": 34445
    },
    {
      "entropy": 5.626059198379517,
      "epoch": 3.125,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040366655874063645,
      "loss": 5.22,
      "mean_token_accuracy": 0.1956747442483902,
      "num_tokens": 71485266.0,
      "step": 34450
    },
    {
      "entropy": 5.505746841430664,
      "epoch": 3.1254535558780843,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004036400135869156,
      "loss": 5.022,
      "mean_token_accuracy": 0.20433289259672166,
      "num_tokens": 71495248.0,
      "step": 34455
    },
    {
      "entropy": 5.545329618453979,
      "epoch": 3.1259071117561685,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040361346577288645,
      "loss": 5.1337,
      "mean_token_accuracy": 0.19765842705965042,
      "num_tokens": 71505600.0,
      "step": 34460
    },
    {
      "entropy": 5.58490481376648,
      "epoch": 3.1263606676342524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040358691529909796,
      "loss": 5.1926,
      "mean_token_accuracy": 0.19311334192752838,
      "num_tokens": 71516634.0,
      "step": 34465
    },
    {
      "entropy": 5.614488124847412,
      "epoch": 3.1268142235123366,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040356036216609943,
      "loss": 5.1586,
      "mean_token_accuracy": 0.1884781613945961,
      "num_tokens": 71526745.0,
      "step": 34470
    },
    {
      "entropy": 5.579267692565918,
      "epoch": 3.127267779390421,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040353380637443974,
      "loss": 5.2736,
      "mean_token_accuracy": 0.18384718149900436,
      "num_tokens": 71537015.0,
      "step": 34475
    },
    {
      "entropy": 5.517903184890747,
      "epoch": 3.127721335268505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040350724792466826,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20211136639118193,
      "num_tokens": 71546660.0,
      "step": 34480
    },
    {
      "entropy": 5.547428369522095,
      "epoch": 3.1281748911465894,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004034806868173341,
      "loss": 5.1669,
      "mean_token_accuracy": 0.19392147064208984,
      "num_tokens": 71557309.0,
      "step": 34485
    },
    {
      "entropy": 5.511440324783325,
      "epoch": 3.1286284470246732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004034541230529866,
      "loss": 5.0708,
      "mean_token_accuracy": 0.19363974928855895,
      "num_tokens": 71567651.0,
      "step": 34490
    },
    {
      "entropy": 5.494543123245239,
      "epoch": 3.1290820029027575,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040342755663217507,
      "loss": 5.0818,
      "mean_token_accuracy": 0.19998909533023834,
      "num_tokens": 71577538.0,
      "step": 34495
    },
    {
      "entropy": 5.521139001846313,
      "epoch": 3.1295355587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000403400987555449,
      "loss": 5.1132,
      "mean_token_accuracy": 0.2005192294716835,
      "num_tokens": 71588299.0,
      "step": 34500
    },
    {
      "entropy": 5.570035171508789,
      "epoch": 3.129989114658926,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004033744158233577,
      "loss": 5.2426,
      "mean_token_accuracy": 0.19168528467416762,
      "num_tokens": 71598989.0,
      "step": 34505
    },
    {
      "entropy": 5.574726343154907,
      "epoch": 3.1304426705370103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040334784143645087,
      "loss": 5.1395,
      "mean_token_accuracy": 0.19435789585113525,
      "num_tokens": 71608596.0,
      "step": 34510
    },
    {
      "entropy": 5.478312206268311,
      "epoch": 3.1308962264150946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004033212643952779,
      "loss": 5.1005,
      "mean_token_accuracy": 0.19651899188756944,
      "num_tokens": 71619182.0,
      "step": 34515
    },
    {
      "entropy": 5.551795578002929,
      "epoch": 3.1313497822931784,
      "grad_norm": 1.25,
      "learning_rate": 0.00040329468470038857,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19439320415258407,
      "num_tokens": 71629526.0,
      "step": 34520
    },
    {
      "entropy": 5.602593994140625,
      "epoch": 3.1318033381712627,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004032681023523324,
      "loss": 5.2228,
      "mean_token_accuracy": 0.1901046007871628,
      "num_tokens": 71640146.0,
      "step": 34525
    },
    {
      "entropy": 5.5074066638946535,
      "epoch": 3.132256894049347,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004032415173516592,
      "loss": 5.0496,
      "mean_token_accuracy": 0.1959142193198204,
      "num_tokens": 71649652.0,
      "step": 34530
    },
    {
      "entropy": 5.512449598312378,
      "epoch": 3.132710449927431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004032149296989187,
      "loss": 5.1202,
      "mean_token_accuracy": 0.20050448328256607,
      "num_tokens": 71659264.0,
      "step": 34535
    },
    {
      "entropy": 5.544788217544555,
      "epoch": 3.133164005805515,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040318833939466083,
      "loss": 5.1559,
      "mean_token_accuracy": 0.19194259345531464,
      "num_tokens": 71669175.0,
      "step": 34540
    },
    {
      "entropy": 5.552178382873535,
      "epoch": 3.1336175616835993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004031617464394354,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1950354740023613,
      "num_tokens": 71679537.0,
      "step": 34545
    },
    {
      "entropy": 5.615230226516724,
      "epoch": 3.1340711175616836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004031351508337925,
      "loss": 5.229,
      "mean_token_accuracy": 0.19516660422086715,
      "num_tokens": 71690024.0,
      "step": 34550
    },
    {
      "entropy": 5.6205860614776615,
      "epoch": 3.134524673439768,
      "grad_norm": 1.375,
      "learning_rate": 0.0004031085525782819,
      "loss": 5.1517,
      "mean_token_accuracy": 0.1962128221988678,
      "num_tokens": 71700019.0,
      "step": 34555
    },
    {
      "entropy": 5.517199325561523,
      "epoch": 3.134978229317852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004030819516734539,
      "loss": 5.184,
      "mean_token_accuracy": 0.18277760744094848,
      "num_tokens": 71710570.0,
      "step": 34560
    },
    {
      "entropy": 5.582555866241455,
      "epoch": 3.1354317851959363,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040305534811985844,
      "loss": 5.2052,
      "mean_token_accuracy": 0.19890635460615158,
      "num_tokens": 71720619.0,
      "step": 34565
    },
    {
      "entropy": 5.5617903709411625,
      "epoch": 3.13588534107402,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004030287419180457,
      "loss": 5.1524,
      "mean_token_accuracy": 0.1962382063269615,
      "num_tokens": 71730201.0,
      "step": 34570
    },
    {
      "entropy": 5.539200162887573,
      "epoch": 3.1363388969521044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040300213306856606,
      "loss": 5.1878,
      "mean_token_accuracy": 0.19693085700273513,
      "num_tokens": 71741368.0,
      "step": 34575
    },
    {
      "entropy": 5.508828449249267,
      "epoch": 3.1367924528301887,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004029755215719696,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19873248487710954,
      "num_tokens": 71751361.0,
      "step": 34580
    },
    {
      "entropy": 5.585836458206177,
      "epoch": 3.137246008708273,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040294890742880677,
      "loss": 5.1374,
      "mean_token_accuracy": 0.19846140891313552,
      "num_tokens": 71761164.0,
      "step": 34585
    },
    {
      "entropy": 5.490891170501709,
      "epoch": 3.1376995645863572,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040292229063962787,
      "loss": 5.1324,
      "mean_token_accuracy": 0.20020836144685744,
      "num_tokens": 71770726.0,
      "step": 34590
    },
    {
      "entropy": 5.55863208770752,
      "epoch": 3.138153120464441,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004028956712049834,
      "loss": 5.2031,
      "mean_token_accuracy": 0.1994895815849304,
      "num_tokens": 71781124.0,
      "step": 34595
    },
    {
      "entropy": 5.526325941085815,
      "epoch": 3.1386066763425253,
      "grad_norm": 1.1875,
      "learning_rate": 0.000402869049125424,
      "loss": 5.1326,
      "mean_token_accuracy": 0.19164952933788298,
      "num_tokens": 71791218.0,
      "step": 34600
    },
    {
      "entropy": 5.568683671951294,
      "epoch": 3.1390602322206096,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004028424244015,
      "loss": 5.1437,
      "mean_token_accuracy": 0.2004018798470497,
      "num_tokens": 71801013.0,
      "step": 34605
    },
    {
      "entropy": 5.5794150829315186,
      "epoch": 3.139513788098694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000402815797033762,
      "loss": 5.2398,
      "mean_token_accuracy": 0.1840790718793869,
      "num_tokens": 71811742.0,
      "step": 34610
    },
    {
      "entropy": 5.476015710830689,
      "epoch": 3.139967343976778,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004027891670227609,
      "loss": 5.1358,
      "mean_token_accuracy": 0.19146006405353547,
      "num_tokens": 71822265.0,
      "step": 34615
    },
    {
      "entropy": 5.543837642669677,
      "epoch": 3.140420899854862,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004027625343690472,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19699826538562776,
      "num_tokens": 71832090.0,
      "step": 34620
    },
    {
      "entropy": 5.528694915771484,
      "epoch": 3.140874455732946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004027358990731717,
      "loss": 5.1163,
      "mean_token_accuracy": 0.1966465786099434,
      "num_tokens": 71841381.0,
      "step": 34625
    },
    {
      "entropy": 5.507441568374634,
      "epoch": 3.1413280116110305,
      "grad_norm": 1.125,
      "learning_rate": 0.00040270926113568524,
      "loss": 5.1627,
      "mean_token_accuracy": 0.1927633300423622,
      "num_tokens": 71851383.0,
      "step": 34630
    },
    {
      "entropy": 5.52910532951355,
      "epoch": 3.1417815674891147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040268262055713873,
      "loss": 5.1534,
      "mean_token_accuracy": 0.19471609443426133,
      "num_tokens": 71862588.0,
      "step": 34635
    },
    {
      "entropy": 5.528946447372436,
      "epoch": 3.142235123367199,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004026559773380831,
      "loss": 5.1356,
      "mean_token_accuracy": 0.1958792582154274,
      "num_tokens": 71873975.0,
      "step": 34640
    },
    {
      "entropy": 5.57400951385498,
      "epoch": 3.142688679245283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004026293314790694,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18359415829181672,
      "num_tokens": 71884084.0,
      "step": 34645
    },
    {
      "entropy": 5.52499737739563,
      "epoch": 3.143142235123367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004026026829806485,
      "loss": 5.1515,
      "mean_token_accuracy": 0.18893705010414125,
      "num_tokens": 71894643.0,
      "step": 34650
    },
    {
      "entropy": 5.489815711975098,
      "epoch": 3.1435957910014514,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040257603184337167,
      "loss": 5.1115,
      "mean_token_accuracy": 0.19511099457740783,
      "num_tokens": 71905245.0,
      "step": 34655
    },
    {
      "entropy": 5.611072635650634,
      "epoch": 3.1440493468795356,
      "grad_norm": 1.5,
      "learning_rate": 0.00040254937806778995,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18545805662870407,
      "num_tokens": 71916473.0,
      "step": 34660
    },
    {
      "entropy": 5.497336721420288,
      "epoch": 3.14450290275762,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040252272165445456,
      "loss": 5.1444,
      "mean_token_accuracy": 0.19085320830345154,
      "num_tokens": 71927465.0,
      "step": 34665
    },
    {
      "entropy": 5.54853310585022,
      "epoch": 3.1449564586357037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004024960626039168,
      "loss": 5.1748,
      "mean_token_accuracy": 0.19498514384031296,
      "num_tokens": 71937380.0,
      "step": 34670
    },
    {
      "entropy": 5.585103511810303,
      "epoch": 3.145410014513788,
      "grad_norm": 1.875,
      "learning_rate": 0.00040246940091672806,
      "loss": 5.2074,
      "mean_token_accuracy": 0.19823941737413406,
      "num_tokens": 71948088.0,
      "step": 34675
    },
    {
      "entropy": 5.541157579421997,
      "epoch": 3.1458635703918723,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004024427365934395,
      "loss": 5.0996,
      "mean_token_accuracy": 0.2021210551261902,
      "num_tokens": 71958284.0,
      "step": 34680
    },
    {
      "entropy": 5.497835731506347,
      "epoch": 3.1463171262699565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040241606963460283,
      "loss": 5.2174,
      "mean_token_accuracy": 0.1886003702878952,
      "num_tokens": 71968138.0,
      "step": 34685
    },
    {
      "entropy": 5.618879652023315,
      "epoch": 3.146770682148041,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040238940004076924,
      "loss": 5.2254,
      "mean_token_accuracy": 0.19073978811502457,
      "num_tokens": 71978275.0,
      "step": 34690
    },
    {
      "entropy": 5.534158325195312,
      "epoch": 3.1472242380261246,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040236272781249047,
      "loss": 5.1908,
      "mean_token_accuracy": 0.19198259562253953,
      "num_tokens": 71988875.0,
      "step": 34695
    },
    {
      "entropy": 5.566407680511475,
      "epoch": 3.147677793904209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040233605295031795,
      "loss": 5.1328,
      "mean_token_accuracy": 0.1993854060769081,
      "num_tokens": 71999560.0,
      "step": 34700
    },
    {
      "entropy": 5.528072309494019,
      "epoch": 3.148131349782293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004023093754548034,
      "loss": 5.1518,
      "mean_token_accuracy": 0.19993666112422942,
      "num_tokens": 72009970.0,
      "step": 34705
    },
    {
      "entropy": 5.537712574005127,
      "epoch": 3.1485849056603774,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004022826953264986,
      "loss": 5.1946,
      "mean_token_accuracy": 0.19959946125745773,
      "num_tokens": 72020196.0,
      "step": 34710
    },
    {
      "entropy": 5.5383553981781,
      "epoch": 3.1490384615384617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040225601256595525,
      "loss": 5.1292,
      "mean_token_accuracy": 0.19098724871873857,
      "num_tokens": 72031456.0,
      "step": 34715
    },
    {
      "entropy": 5.565259599685669,
      "epoch": 3.149492017416546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000402229327173725,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1920481100678444,
      "num_tokens": 72042260.0,
      "step": 34720
    },
    {
      "entropy": 5.620397758483887,
      "epoch": 3.1499455732946298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040220263915035997,
      "loss": 5.2844,
      "mean_token_accuracy": 0.17728457748889923,
      "num_tokens": 72053463.0,
      "step": 34725
    },
    {
      "entropy": 5.66829514503479,
      "epoch": 3.150399129172714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040217594849641195,
      "loss": 5.2882,
      "mean_token_accuracy": 0.18419014364480973,
      "num_tokens": 72064508.0,
      "step": 34730
    },
    {
      "entropy": 5.506241512298584,
      "epoch": 3.1508526850507983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040214925521243287,
      "loss": 5.1722,
      "mean_token_accuracy": 0.18866872936487197,
      "num_tokens": 72074758.0,
      "step": 34735
    },
    {
      "entropy": 5.511781883239746,
      "epoch": 3.1513062409288826,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040212255929897474,
      "loss": 5.107,
      "mean_token_accuracy": 0.19962465912103652,
      "num_tokens": 72085548.0,
      "step": 34740
    },
    {
      "entropy": 5.557574129104614,
      "epoch": 3.1517597968069664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004020958607565897,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18943857550621032,
      "num_tokens": 72096773.0,
      "step": 34745
    },
    {
      "entropy": 5.542710161209106,
      "epoch": 3.1522133526850507,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004020691595858299,
      "loss": 5.2133,
      "mean_token_accuracy": 0.19783455282449722,
      "num_tokens": 72106832.0,
      "step": 34750
    },
    {
      "entropy": 5.568316268920898,
      "epoch": 3.152666908563135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040204245578724754,
      "loss": 5.2111,
      "mean_token_accuracy": 0.19509605765342714,
      "num_tokens": 72117758.0,
      "step": 34755
    },
    {
      "entropy": 5.594624471664429,
      "epoch": 3.153120464441219,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040201574936139475,
      "loss": 5.1177,
      "mean_token_accuracy": 0.1927069365978241,
      "num_tokens": 72127522.0,
      "step": 34760
    },
    {
      "entropy": 5.484645223617553,
      "epoch": 3.1535740203193035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004019890403088239,
      "loss": 5.1701,
      "mean_token_accuracy": 0.19670455157756805,
      "num_tokens": 72138310.0,
      "step": 34765
    },
    {
      "entropy": 5.5425842761993405,
      "epoch": 3.1540275761973877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040196232863008735,
      "loss": 5.1982,
      "mean_token_accuracy": 0.20032694637775422,
      "num_tokens": 72148958.0,
      "step": 34770
    },
    {
      "entropy": 5.620175552368164,
      "epoch": 3.1544811320754715,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040193561432573754,
      "loss": 5.1558,
      "mean_token_accuracy": 0.19763898104429245,
      "num_tokens": 72158594.0,
      "step": 34775
    },
    {
      "entropy": 5.577758836746216,
      "epoch": 3.154934687953556,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004019088973963268,
      "loss": 5.2338,
      "mean_token_accuracy": 0.19159920066595076,
      "num_tokens": 72168552.0,
      "step": 34780
    },
    {
      "entropy": 5.537615060806274,
      "epoch": 3.15538824383164,
      "grad_norm": 2.140625,
      "learning_rate": 0.00040188217784240785,
      "loss": 5.1054,
      "mean_token_accuracy": 0.19777926951646804,
      "num_tokens": 72178584.0,
      "step": 34785
    },
    {
      "entropy": 5.506344270706177,
      "epoch": 3.1558417997097243,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000401855455664533,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1992138534784317,
      "num_tokens": 72189144.0,
      "step": 34790
    },
    {
      "entropy": 5.596024799346924,
      "epoch": 3.1562953555878086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004018287308632551,
      "loss": 5.1695,
      "mean_token_accuracy": 0.19470119327306748,
      "num_tokens": 72198326.0,
      "step": 34795
    },
    {
      "entropy": 5.5687438488006595,
      "epoch": 3.1567489114658924,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004018020034391266,
      "loss": 5.1545,
      "mean_token_accuracy": 0.19860903471708297,
      "num_tokens": 72207916.0,
      "step": 34800
    },
    {
      "entropy": 5.613378810882568,
      "epoch": 3.1572024673439767,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004017752733927004,
      "loss": 5.1946,
      "mean_token_accuracy": 0.19678377956151963,
      "num_tokens": 72219111.0,
      "step": 34805
    },
    {
      "entropy": 5.513866710662842,
      "epoch": 3.157656023222061,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004017485407245293,
      "loss": 5.1924,
      "mean_token_accuracy": 0.20028549283742905,
      "num_tokens": 72230081.0,
      "step": 34810
    },
    {
      "entropy": 5.479847383499146,
      "epoch": 3.1581095791001452,
      "grad_norm": 1.21875,
      "learning_rate": 0.000401721805435166,
      "loss": 5.1316,
      "mean_token_accuracy": 0.1964292496442795,
      "num_tokens": 72240693.0,
      "step": 34815
    },
    {
      "entropy": 5.505267238616943,
      "epoch": 3.1585631349782295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040169506752516356,
      "loss": 5.0887,
      "mean_token_accuracy": 0.20486992746591567,
      "num_tokens": 72250925.0,
      "step": 34820
    },
    {
      "entropy": 5.484885931015015,
      "epoch": 3.1590166908563133,
      "grad_norm": 1.125,
      "learning_rate": 0.0004016683269950748,
      "loss": 5.0821,
      "mean_token_accuracy": 0.20893379747867585,
      "num_tokens": 72261027.0,
      "step": 34825
    },
    {
      "entropy": 5.528301048278808,
      "epoch": 3.1594702467343976,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040164158384545267,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2150461956858635,
      "num_tokens": 72269430.0,
      "step": 34830
    },
    {
      "entropy": 5.438672399520874,
      "epoch": 3.159923802612482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004016148380768504,
      "loss": 5.0809,
      "mean_token_accuracy": 0.20243236869573594,
      "num_tokens": 72280207.0,
      "step": 34835
    },
    {
      "entropy": 5.542630863189697,
      "epoch": 3.160377358490566,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040158808968982103,
      "loss": 5.269,
      "mean_token_accuracy": 0.19229751229286193,
      "num_tokens": 72290093.0,
      "step": 34840
    },
    {
      "entropy": 5.571558856964112,
      "epoch": 3.1608309143686504,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040156133868491766,
      "loss": 5.0923,
      "mean_token_accuracy": 0.1994951456785202,
      "num_tokens": 72300203.0,
      "step": 34845
    },
    {
      "entropy": 5.557386684417724,
      "epoch": 3.161284470246734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040153458506269365,
      "loss": 5.1826,
      "mean_token_accuracy": 0.19029328972101212,
      "num_tokens": 72311778.0,
      "step": 34850
    },
    {
      "entropy": 5.554348182678223,
      "epoch": 3.1617380261248185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000401507828823702,
      "loss": 5.147,
      "mean_token_accuracy": 0.1922489270567894,
      "num_tokens": 72322142.0,
      "step": 34855
    },
    {
      "entropy": 5.549812173843383,
      "epoch": 3.1621915820029027,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040148106996849636,
      "loss": 5.2214,
      "mean_token_accuracy": 0.19350111186504365,
      "num_tokens": 72332312.0,
      "step": 34860
    },
    {
      "entropy": 5.452987909317017,
      "epoch": 3.162645137880987,
      "grad_norm": 1.25,
      "learning_rate": 0.00040145430849762993,
      "loss": 5.0938,
      "mean_token_accuracy": 0.19451790004968644,
      "num_tokens": 72342157.0,
      "step": 34865
    },
    {
      "entropy": 5.607350587844849,
      "epoch": 3.1630986937590713,
      "grad_norm": 1.5625,
      "learning_rate": 0.00040142754441165615,
      "loss": 5.1696,
      "mean_token_accuracy": 0.1966680720448494,
      "num_tokens": 72353417.0,
      "step": 34870
    },
    {
      "entropy": 5.63164381980896,
      "epoch": 3.1635522496371555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004014007777111285,
      "loss": 5.142,
      "mean_token_accuracy": 0.1981915071606636,
      "num_tokens": 72363744.0,
      "step": 34875
    },
    {
      "entropy": 5.551121711730957,
      "epoch": 3.1640058055152394,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040137400839660066,
      "loss": 5.2045,
      "mean_token_accuracy": 0.1931806594133377,
      "num_tokens": 72375241.0,
      "step": 34880
    },
    {
      "entropy": 5.556085729598999,
      "epoch": 3.1644593613933236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040134723646862607,
      "loss": 5.1673,
      "mean_token_accuracy": 0.1970580130815506,
      "num_tokens": 72385475.0,
      "step": 34885
    },
    {
      "entropy": 5.544161367416382,
      "epoch": 3.164912917271408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040132046192775835,
      "loss": 5.2354,
      "mean_token_accuracy": 0.18955578804016113,
      "num_tokens": 72396417.0,
      "step": 34890
    },
    {
      "entropy": 5.562213325500489,
      "epoch": 3.165366473149492,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004012936847745514,
      "loss": 5.1844,
      "mean_token_accuracy": 0.193455071747303,
      "num_tokens": 72406169.0,
      "step": 34895
    },
    {
      "entropy": 5.582019090652466,
      "epoch": 3.165820029027576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004012669050095589,
      "loss": 5.1831,
      "mean_token_accuracy": 0.19047532975673676,
      "num_tokens": 72416522.0,
      "step": 34900
    },
    {
      "entropy": 5.571487331390381,
      "epoch": 3.1662735849056602,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004012401226333345,
      "loss": 5.2063,
      "mean_token_accuracy": 0.1949528932571411,
      "num_tokens": 72426000.0,
      "step": 34905
    },
    {
      "entropy": 5.603312969207764,
      "epoch": 3.1667271407837445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004012133376464322,
      "loss": 5.2538,
      "mean_token_accuracy": 0.18847629725933074,
      "num_tokens": 72435952.0,
      "step": 34910
    },
    {
      "entropy": 5.596228456497192,
      "epoch": 3.1671806966618288,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00040118655004940597,
      "loss": 5.2049,
      "mean_token_accuracy": 0.19271286875009536,
      "num_tokens": 72449957.0,
      "step": 34915
    },
    {
      "entropy": 5.593516874313354,
      "epoch": 3.167634252539913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004011597598428097,
      "loss": 5.1933,
      "mean_token_accuracy": 0.19133229851722716,
      "num_tokens": 72459956.0,
      "step": 34920
    },
    {
      "entropy": 5.541787719726562,
      "epoch": 3.1680878084179973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004011329670271975,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19942363798618318,
      "num_tokens": 72471107.0,
      "step": 34925
    },
    {
      "entropy": 5.448157787322998,
      "epoch": 3.168541364296081,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040110617160312334,
      "loss": 5.1016,
      "mean_token_accuracy": 0.20398622155189514,
      "num_tokens": 72481428.0,
      "step": 34930
    },
    {
      "entropy": 5.571733856201172,
      "epoch": 3.1689949201741654,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004010793735711415,
      "loss": 5.23,
      "mean_token_accuracy": 0.18585646003484727,
      "num_tokens": 72491133.0,
      "step": 34935
    },
    {
      "entropy": 5.532226610183716,
      "epoch": 3.1694484760522497,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040105257293180605,
      "loss": 5.1023,
      "mean_token_accuracy": 0.2021437406539917,
      "num_tokens": 72502047.0,
      "step": 34940
    },
    {
      "entropy": 5.54587607383728,
      "epoch": 3.169902031930334,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004010257696856712,
      "loss": 5.1753,
      "mean_token_accuracy": 0.18797962367534637,
      "num_tokens": 72513120.0,
      "step": 34945
    },
    {
      "entropy": 5.572798585891723,
      "epoch": 3.170355587808418,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00040099896383329135,
      "loss": 5.26,
      "mean_token_accuracy": 0.19841453433036804,
      "num_tokens": 72523118.0,
      "step": 34950
    },
    {
      "entropy": 5.624402475357056,
      "epoch": 3.170809143686502,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040097215537522086,
      "loss": 5.2834,
      "mean_token_accuracy": 0.19168690741062164,
      "num_tokens": 72533697.0,
      "step": 34955
    },
    {
      "entropy": 5.53308744430542,
      "epoch": 3.1712626995645863,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040094534431201406,
      "loss": 5.0909,
      "mean_token_accuracy": 0.2048969864845276,
      "num_tokens": 72543661.0,
      "step": 34960
    },
    {
      "entropy": 5.602494859695435,
      "epoch": 3.1717162554426706,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004009185306442255,
      "loss": 5.258,
      "mean_token_accuracy": 0.1899183601140976,
      "num_tokens": 72554551.0,
      "step": 34965
    },
    {
      "entropy": 5.585455751419067,
      "epoch": 3.172169811320755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040089171437240954,
      "loss": 5.2516,
      "mean_token_accuracy": 0.18665444254875183,
      "num_tokens": 72565134.0,
      "step": 34970
    },
    {
      "entropy": 5.541125726699829,
      "epoch": 3.172623367198839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040086489549712097,
      "loss": 5.1627,
      "mean_token_accuracy": 0.19593857526779174,
      "num_tokens": 72574981.0,
      "step": 34975
    },
    {
      "entropy": 5.521917104721069,
      "epoch": 3.173076923076923,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004008380740189142,
      "loss": 5.1646,
      "mean_token_accuracy": 0.19387368559837342,
      "num_tokens": 72585073.0,
      "step": 34980
    },
    {
      "entropy": 5.564421081542969,
      "epoch": 3.173530478955007,
      "grad_norm": 1.765625,
      "learning_rate": 0.000400811249938344,
      "loss": 5.1863,
      "mean_token_accuracy": 0.1974617525935173,
      "num_tokens": 72595719.0,
      "step": 34985
    },
    {
      "entropy": 5.559382581710816,
      "epoch": 3.1739840348330914,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004007844232559652,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1915575459599495,
      "num_tokens": 72605276.0,
      "step": 34990
    },
    {
      "entropy": 5.6596063613891605,
      "epoch": 3.1744375907111757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004007575939723324,
      "loss": 5.2119,
      "mean_token_accuracy": 0.190667088329792,
      "num_tokens": 72614964.0,
      "step": 34995
    },
    {
      "entropy": 5.645944976806641,
      "epoch": 3.17489114658926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004007307620880004,
      "loss": 5.1996,
      "mean_token_accuracy": 0.19533899575471877,
      "num_tokens": 72625461.0,
      "step": 35000
    },
    {
      "entropy": 5.530227375030518,
      "epoch": 3.175344702467344,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040070392760352424,
      "loss": 5.2061,
      "mean_token_accuracy": 0.1943572849035263,
      "num_tokens": 72636192.0,
      "step": 35005
    },
    {
      "entropy": 5.5364768505096436,
      "epoch": 3.175798258345428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040067709051945895,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1910842627286911,
      "num_tokens": 72648432.0,
      "step": 35010
    },
    {
      "entropy": 5.552248859405518,
      "epoch": 3.1762518142235123,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004006502508363593,
      "loss": 5.1379,
      "mean_token_accuracy": 0.19481161683797837,
      "num_tokens": 72658678.0,
      "step": 35015
    },
    {
      "entropy": 5.648138856887817,
      "epoch": 3.1767053701015966,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040062340855478046,
      "loss": 5.3483,
      "mean_token_accuracy": 0.18174565434455872,
      "num_tokens": 72669191.0,
      "step": 35020
    },
    {
      "entropy": 5.5978985786437985,
      "epoch": 3.177158925979681,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040059656367527754,
      "loss": 5.1367,
      "mean_token_accuracy": 0.19402617812156678,
      "num_tokens": 72679327.0,
      "step": 35025
    },
    {
      "entropy": 5.55527024269104,
      "epoch": 3.1776124818577647,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004005697161984056,
      "loss": 5.2188,
      "mean_token_accuracy": 0.1912990018725395,
      "num_tokens": 72689673.0,
      "step": 35030
    },
    {
      "entropy": 5.512010765075684,
      "epoch": 3.178066037735849,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040054286612472006,
      "loss": 5.2251,
      "mean_token_accuracy": 0.18525668680667878,
      "num_tokens": 72700606.0,
      "step": 35035
    },
    {
      "entropy": 5.613732242584229,
      "epoch": 3.178519593613933,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040051601345477593,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19464820325374604,
      "num_tokens": 72711195.0,
      "step": 35040
    },
    {
      "entropy": 5.547444677352905,
      "epoch": 3.1789731494920175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040048915818912875,
      "loss": 5.1845,
      "mean_token_accuracy": 0.1951794981956482,
      "num_tokens": 72721160.0,
      "step": 35045
    },
    {
      "entropy": 5.532799959182739,
      "epoch": 3.1794267053701017,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040046230032833374,
      "loss": 5.1857,
      "mean_token_accuracy": 0.19116449356079102,
      "num_tokens": 72730867.0,
      "step": 35050
    },
    {
      "entropy": 5.538054370880127,
      "epoch": 3.1798802612481856,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004004354398729464,
      "loss": 5.1803,
      "mean_token_accuracy": 0.19359404891729354,
      "num_tokens": 72740453.0,
      "step": 35055
    },
    {
      "entropy": 5.612128114700317,
      "epoch": 3.18033381712627,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004004085768235222,
      "loss": 5.219,
      "mean_token_accuracy": 0.18665474206209182,
      "num_tokens": 72751395.0,
      "step": 35060
    },
    {
      "entropy": 5.569827890396118,
      "epoch": 3.180787373004354,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040038171118061673,
      "loss": 5.1442,
      "mean_token_accuracy": 0.19463087618350983,
      "num_tokens": 72761300.0,
      "step": 35065
    },
    {
      "entropy": 5.526641750335694,
      "epoch": 3.1812409288824384,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004003548429447855,
      "loss": 5.1469,
      "mean_token_accuracy": 0.2044302389025688,
      "num_tokens": 72772011.0,
      "step": 35070
    },
    {
      "entropy": 5.5373451709747314,
      "epoch": 3.1816944847605226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004003279721165841,
      "loss": 5.1887,
      "mean_token_accuracy": 0.18980462849140167,
      "num_tokens": 72782521.0,
      "step": 35075
    },
    {
      "entropy": 5.482511043548584,
      "epoch": 3.182148040638607,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004003010986965683,
      "loss": 5.064,
      "mean_token_accuracy": 0.20144071727991103,
      "num_tokens": 72793125.0,
      "step": 35080
    },
    {
      "entropy": 5.61639928817749,
      "epoch": 3.1826015965166907,
      "grad_norm": 1.34375,
      "learning_rate": 0.000400274222685294,
      "loss": 5.2275,
      "mean_token_accuracy": 0.18940045088529586,
      "num_tokens": 72804541.0,
      "step": 35085
    },
    {
      "entropy": 5.710501718521118,
      "epoch": 3.183055152394775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040024734408331664,
      "loss": 5.287,
      "mean_token_accuracy": 0.19319023340940475,
      "num_tokens": 72814447.0,
      "step": 35090
    },
    {
      "entropy": 5.5708812236785885,
      "epoch": 3.1835087082728593,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040022046289119246,
      "loss": 5.2047,
      "mean_token_accuracy": 0.19861464202404022,
      "num_tokens": 72823545.0,
      "step": 35095
    },
    {
      "entropy": 5.55828332901001,
      "epoch": 3.1839622641509435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040019357910947714,
      "loss": 5.091,
      "mean_token_accuracy": 0.19695786386728287,
      "num_tokens": 72833803.0,
      "step": 35100
    },
    {
      "entropy": 5.581221199035644,
      "epoch": 3.1844158200290273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040016669273872666,
      "loss": 5.1835,
      "mean_token_accuracy": 0.19660593420267106,
      "num_tokens": 72843383.0,
      "step": 35105
    },
    {
      "entropy": 5.533212089538575,
      "epoch": 3.1848693759071116,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040013980377949716,
      "loss": 5.2205,
      "mean_token_accuracy": 0.19096733182668685,
      "num_tokens": 72853285.0,
      "step": 35110
    },
    {
      "entropy": 5.5772451877594,
      "epoch": 3.185322931785196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040011291223234456,
      "loss": 5.2199,
      "mean_token_accuracy": 0.1932971239089966,
      "num_tokens": 72863556.0,
      "step": 35115
    },
    {
      "entropy": 5.577397346496582,
      "epoch": 3.18577648766328,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040008601809782507,
      "loss": 5.1124,
      "mean_token_accuracy": 0.1956990987062454,
      "num_tokens": 72873296.0,
      "step": 35120
    },
    {
      "entropy": 5.577636575698852,
      "epoch": 3.1862300435413644,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004000591213764948,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18740827441215516,
      "num_tokens": 72884738.0,
      "step": 35125
    },
    {
      "entropy": 5.595229244232177,
      "epoch": 3.1866835994194487,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040003222206891017,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18874747306108475,
      "num_tokens": 72894573.0,
      "step": 35130
    },
    {
      "entropy": 5.52909107208252,
      "epoch": 3.1871371552975325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004000053201756272,
      "loss": 5.1124,
      "mean_token_accuracy": 0.19802445471286773,
      "num_tokens": 72904976.0,
      "step": 35135
    },
    {
      "entropy": 5.54230432510376,
      "epoch": 3.1875907111756168,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003999784156972024,
      "loss": 5.1827,
      "mean_token_accuracy": 0.19015426337718963,
      "num_tokens": 72915152.0,
      "step": 35140
    },
    {
      "entropy": 5.589487838745117,
      "epoch": 3.188044267053701,
      "grad_norm": 1.75,
      "learning_rate": 0.0003999515086341921,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18871065080165864,
      "num_tokens": 72925450.0,
      "step": 35145
    },
    {
      "entropy": 5.524220132827759,
      "epoch": 3.1884978229317853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003999245989871527,
      "loss": 5.0983,
      "mean_token_accuracy": 0.19713550955057144,
      "num_tokens": 72935990.0,
      "step": 35150
    },
    {
      "entropy": 5.513463163375855,
      "epoch": 3.1889513788098696,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00039989768675664084,
      "loss": 5.0477,
      "mean_token_accuracy": 0.1952058881521225,
      "num_tokens": 72945662.0,
      "step": 35155
    },
    {
      "entropy": 5.481577348709107,
      "epoch": 3.1894049346879534,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039987077194321305,
      "loss": 5.001,
      "mean_token_accuracy": 0.20281193107366563,
      "num_tokens": 72955313.0,
      "step": 35160
    },
    {
      "entropy": 5.597303342819214,
      "epoch": 3.1898584905660377,
      "grad_norm": 1.578125,
      "learning_rate": 0.00039984385454742576,
      "loss": 5.2709,
      "mean_token_accuracy": 0.1864163503050804,
      "num_tokens": 72965347.0,
      "step": 35165
    },
    {
      "entropy": 5.568429565429687,
      "epoch": 3.190312046444122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039981693456983574,
      "loss": 5.1081,
      "mean_token_accuracy": 0.19588731229305267,
      "num_tokens": 72976794.0,
      "step": 35170
    },
    {
      "entropy": 5.519783592224121,
      "epoch": 3.190765602322206,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003997900120109998,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2009709060192108,
      "num_tokens": 72986837.0,
      "step": 35175
    },
    {
      "entropy": 5.5039159774780275,
      "epoch": 3.1912191582002905,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003997630868714745,
      "loss": 5.137,
      "mean_token_accuracy": 0.1915770649909973,
      "num_tokens": 72996819.0,
      "step": 35180
    },
    {
      "entropy": 5.444828796386719,
      "epoch": 3.1916727140783743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039973615915181687,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20517793148756028,
      "num_tokens": 73005963.0,
      "step": 35185
    },
    {
      "entropy": 5.488290643692016,
      "epoch": 3.1921262699564585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039970922885258363,
      "loss": 5.1935,
      "mean_token_accuracy": 0.1898084506392479,
      "num_tokens": 73015912.0,
      "step": 35190
    },
    {
      "entropy": 5.5121180534362795,
      "epoch": 3.192579825834543,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003996822959743318,
      "loss": 5.0619,
      "mean_token_accuracy": 0.20445189327001573,
      "num_tokens": 73026528.0,
      "step": 35195
    },
    {
      "entropy": 5.555868530273438,
      "epoch": 3.193033381712627,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003996553605176183,
      "loss": 5.1936,
      "mean_token_accuracy": 0.19174249172210694,
      "num_tokens": 73036255.0,
      "step": 35200
    },
    {
      "entropy": 5.575733757019043,
      "epoch": 3.1934869375907113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003996284224830002,
      "loss": 5.2889,
      "mean_token_accuracy": 0.19036538451910018,
      "num_tokens": 73046686.0,
      "step": 35205
    },
    {
      "entropy": 5.533800983428955,
      "epoch": 3.193940493468795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039960148187103453,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20139340907335282,
      "num_tokens": 73056070.0,
      "step": 35210
    },
    {
      "entropy": 5.6314788341522215,
      "epoch": 3.1943940493468794,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039957453868227845,
      "loss": 5.2053,
      "mean_token_accuracy": 0.1965641051530838,
      "num_tokens": 73066288.0,
      "step": 35215
    },
    {
      "entropy": 5.522759866714478,
      "epoch": 3.1948476052249637,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00039954759291728915,
      "loss": 5.1671,
      "mean_token_accuracy": 0.18972538709640502,
      "num_tokens": 73076297.0,
      "step": 35220
    },
    {
      "entropy": 5.562909030914307,
      "epoch": 3.195301161103048,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039952064457662395,
      "loss": 5.2166,
      "mean_token_accuracy": 0.19347635954618453,
      "num_tokens": 73086249.0,
      "step": 35225
    },
    {
      "entropy": 5.515375804901123,
      "epoch": 3.1957547169811322,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039949369366084,
      "loss": 5.1048,
      "mean_token_accuracy": 0.20154260098934174,
      "num_tokens": 73096358.0,
      "step": 35230
    },
    {
      "entropy": 5.509189414978027,
      "epoch": 3.196208272859216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039946674017049484,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19807538986206055,
      "num_tokens": 73106891.0,
      "step": 35235
    },
    {
      "entropy": 5.458211946487427,
      "epoch": 3.1966618287373003,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039943978410614565,
      "loss": 5.0806,
      "mean_token_accuracy": 0.19734696596860885,
      "num_tokens": 73117106.0,
      "step": 35240
    },
    {
      "entropy": 5.5325061798095705,
      "epoch": 3.1971153846153846,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00039941282546835017,
      "loss": 5.1831,
      "mean_token_accuracy": 0.1957603469491005,
      "num_tokens": 73127112.0,
      "step": 35245
    },
    {
      "entropy": 5.555215501785279,
      "epoch": 3.197568940493469,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039938586425766564,
      "loss": 5.1718,
      "mean_token_accuracy": 0.19060900658369065,
      "num_tokens": 73137751.0,
      "step": 35250
    },
    {
      "entropy": 5.591100788116455,
      "epoch": 3.198022496371553,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003993589004746498,
      "loss": 5.1662,
      "mean_token_accuracy": 0.2002313420176506,
      "num_tokens": 73149303.0,
      "step": 35255
    },
    {
      "entropy": 5.508702945709229,
      "epoch": 3.198476052249637,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039933193411986024,
      "loss": 5.1527,
      "mean_token_accuracy": 0.19114989638328553,
      "num_tokens": 73160025.0,
      "step": 35260
    },
    {
      "entropy": 5.545434045791626,
      "epoch": 3.198929608127721,
      "grad_norm": 2.0625,
      "learning_rate": 0.00039930496519385454,
      "loss": 5.1294,
      "mean_token_accuracy": 0.2029668375849724,
      "num_tokens": 73170721.0,
      "step": 35265
    },
    {
      "entropy": 5.552112245559693,
      "epoch": 3.1993831640058055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039927799369719047,
      "loss": 5.1978,
      "mean_token_accuracy": 0.19069015979766846,
      "num_tokens": 73180902.0,
      "step": 35270
    },
    {
      "entropy": 5.437372922897339,
      "epoch": 3.1998367198838897,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003992510196304258,
      "loss": 5.0103,
      "mean_token_accuracy": 0.2061469092965126,
      "num_tokens": 73191259.0,
      "step": 35275
    },
    {
      "entropy": 5.532780313491822,
      "epoch": 3.200290275761974,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00039922404299411846,
      "loss": 5.1462,
      "mean_token_accuracy": 0.20410795658826827,
      "num_tokens": 73201654.0,
      "step": 35280
    },
    {
      "entropy": 5.5577840328216555,
      "epoch": 3.2007438316400583,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003991970637888262,
      "loss": 5.2106,
      "mean_token_accuracy": 0.1828935757279396,
      "num_tokens": 73212658.0,
      "step": 35285
    },
    {
      "entropy": 5.5681201934814455,
      "epoch": 3.201197387518142,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000399170082015107,
      "loss": 5.1825,
      "mean_token_accuracy": 0.1924119085073471,
      "num_tokens": 73222883.0,
      "step": 35290
    },
    {
      "entropy": 5.523045635223388,
      "epoch": 3.2016509433962264,
      "grad_norm": 1.15625,
      "learning_rate": 0.000399143097673519,
      "loss": 5.106,
      "mean_token_accuracy": 0.1954590857028961,
      "num_tokens": 73233731.0,
      "step": 35295
    },
    {
      "entropy": 5.523660469055176,
      "epoch": 3.2021044992743106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039911611076461995,
      "loss": 5.0654,
      "mean_token_accuracy": 0.20111292451620102,
      "num_tokens": 73243794.0,
      "step": 35300
    },
    {
      "entropy": 5.533521461486816,
      "epoch": 3.202558055152395,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039908912128896817,
      "loss": 5.1394,
      "mean_token_accuracy": 0.18726421892642975,
      "num_tokens": 73254708.0,
      "step": 35305
    },
    {
      "entropy": 5.532264471054077,
      "epoch": 3.203011611030479,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003990621292471217,
      "loss": 5.1111,
      "mean_token_accuracy": 0.1936648517847061,
      "num_tokens": 73265311.0,
      "step": 35310
    },
    {
      "entropy": 5.63657398223877,
      "epoch": 3.203465166908563,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039903513463963874,
      "loss": 5.3288,
      "mean_token_accuracy": 0.19080508053302764,
      "num_tokens": 73275886.0,
      "step": 35315
    },
    {
      "entropy": 5.603589963912964,
      "epoch": 3.2039187227866472,
      "grad_norm": 1.5,
      "learning_rate": 0.0003990081374670777,
      "loss": 5.1817,
      "mean_token_accuracy": 0.19605736881494523,
      "num_tokens": 73285903.0,
      "step": 35320
    },
    {
      "entropy": 5.4861725807189945,
      "epoch": 3.2043722786647315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039898113772999665,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19484651386737822,
      "num_tokens": 73296571.0,
      "step": 35325
    },
    {
      "entropy": 5.56073637008667,
      "epoch": 3.2048258345428158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039895413542895427,
      "loss": 5.1954,
      "mean_token_accuracy": 0.1906631827354431,
      "num_tokens": 73307105.0,
      "step": 35330
    },
    {
      "entropy": 5.498159027099609,
      "epoch": 3.2052793904209,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003989271305645086,
      "loss": 5.1041,
      "mean_token_accuracy": 0.20001456141471863,
      "num_tokens": 73316664.0,
      "step": 35335
    },
    {
      "entropy": 5.518439388275146,
      "epoch": 3.205732946298984,
      "grad_norm": 1.0,
      "learning_rate": 0.00039890012313721835,
      "loss": 5.1837,
      "mean_token_accuracy": 0.20105521231889725,
      "num_tokens": 73327438.0,
      "step": 35340
    },
    {
      "entropy": 5.525814628601074,
      "epoch": 3.206186502177068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000398873113147642,
      "loss": 5.1564,
      "mean_token_accuracy": 0.1998247265815735,
      "num_tokens": 73337635.0,
      "step": 35345
    },
    {
      "entropy": 5.528606271743774,
      "epoch": 3.2066400580551524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039884610059633807,
      "loss": 5.1454,
      "mean_token_accuracy": 0.20092587321996688,
      "num_tokens": 73347032.0,
      "step": 35350
    },
    {
      "entropy": 5.59629635810852,
      "epoch": 3.2070936139332367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003988190854838652,
      "loss": 5.2415,
      "mean_token_accuracy": 0.19119194000959397,
      "num_tokens": 73357780.0,
      "step": 35355
    },
    {
      "entropy": 5.609995651245117,
      "epoch": 3.207547169811321,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039879206781078214,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19606900066137314,
      "num_tokens": 73367831.0,
      "step": 35360
    },
    {
      "entropy": 5.57894549369812,
      "epoch": 3.2080007256894048,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003987650475776476,
      "loss": 5.1671,
      "mean_token_accuracy": 0.1938989371061325,
      "num_tokens": 73378588.0,
      "step": 35365
    },
    {
      "entropy": 5.463727140426636,
      "epoch": 3.208454281567489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039873802478502027,
      "loss": 5.1354,
      "mean_token_accuracy": 0.20116665214300156,
      "num_tokens": 73388414.0,
      "step": 35370
    },
    {
      "entropy": 5.545647001266479,
      "epoch": 3.2089078374455733,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003987109994334591,
      "loss": 5.1633,
      "mean_token_accuracy": 0.19157373607158662,
      "num_tokens": 73398891.0,
      "step": 35375
    },
    {
      "entropy": 5.580695772171021,
      "epoch": 3.2093613933236576,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003986839715235229,
      "loss": 5.2119,
      "mean_token_accuracy": 0.19012885987758638,
      "num_tokens": 73409605.0,
      "step": 35380
    },
    {
      "entropy": 5.590838479995727,
      "epoch": 3.209814949201742,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039865694105577063,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18895650207996367,
      "num_tokens": 73419103.0,
      "step": 35385
    },
    {
      "entropy": 5.428395128250122,
      "epoch": 3.2102685050798256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003986299080307614,
      "loss": 5.1548,
      "mean_token_accuracy": 0.19890921264886857,
      "num_tokens": 73429134.0,
      "step": 35390
    },
    {
      "entropy": 5.585702037811279,
      "epoch": 3.21072206095791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003986028724490541,
      "loss": 5.2189,
      "mean_token_accuracy": 0.18873873502016067,
      "num_tokens": 73440382.0,
      "step": 35395
    },
    {
      "entropy": 5.598724508285523,
      "epoch": 3.211175616835994,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003985758343112079,
      "loss": 5.159,
      "mean_token_accuracy": 0.20106744021177292,
      "num_tokens": 73451300.0,
      "step": 35400
    },
    {
      "entropy": 5.4963384628295895,
      "epoch": 3.2116291727140784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039854879361778193,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1978398457169533,
      "num_tokens": 73461339.0,
      "step": 35405
    },
    {
      "entropy": 5.488100290298462,
      "epoch": 3.2120827285921627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039852175036933545,
      "loss": 5.1576,
      "mean_token_accuracy": 0.1963038206100464,
      "num_tokens": 73470961.0,
      "step": 35410
    },
    {
      "entropy": 5.607589530944824,
      "epoch": 3.2125362844702465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039849470456642777,
      "loss": 5.2324,
      "mean_token_accuracy": 0.18865307569503784,
      "num_tokens": 73481492.0,
      "step": 35415
    },
    {
      "entropy": 5.532109212875366,
      "epoch": 3.212989840348331,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039846765620961806,
      "loss": 5.107,
      "mean_token_accuracy": 0.19386760741472245,
      "num_tokens": 73492351.0,
      "step": 35420
    },
    {
      "entropy": 5.55044264793396,
      "epoch": 3.213443396226415,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003984406052994658,
      "loss": 5.229,
      "mean_token_accuracy": 0.19120626002550126,
      "num_tokens": 73503265.0,
      "step": 35425
    },
    {
      "entropy": 5.556250190734863,
      "epoch": 3.2138969521044993,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003984135518365303,
      "loss": 5.1888,
      "mean_token_accuracy": 0.19025546908378602,
      "num_tokens": 73513804.0,
      "step": 35430
    },
    {
      "entropy": 5.572612285614014,
      "epoch": 3.2143505079825836,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039838649582137113,
      "loss": 5.1848,
      "mean_token_accuracy": 0.19527635872364044,
      "num_tokens": 73523453.0,
      "step": 35435
    },
    {
      "entropy": 5.579759407043457,
      "epoch": 3.214804063860668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039835943725454777,
      "loss": 5.2962,
      "mean_token_accuracy": 0.1852312758564949,
      "num_tokens": 73533931.0,
      "step": 35440
    },
    {
      "entropy": 5.6241058826446535,
      "epoch": 3.2152576197387517,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039833237613661984,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1964004546403885,
      "num_tokens": 73544562.0,
      "step": 35445
    },
    {
      "entropy": 5.547450208663941,
      "epoch": 3.215711175616836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003983053124681469,
      "loss": 5.1566,
      "mean_token_accuracy": 0.20164919048547744,
      "num_tokens": 73555149.0,
      "step": 35450
    },
    {
      "entropy": 5.493491363525391,
      "epoch": 3.21616473149492,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039827824624968873,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19425561279058456,
      "num_tokens": 73565131.0,
      "step": 35455
    },
    {
      "entropy": 5.5084349632263185,
      "epoch": 3.2166182873730045,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039825117748180493,
      "loss": 5.1176,
      "mean_token_accuracy": 0.1997495099902153,
      "num_tokens": 73574185.0,
      "step": 35460
    },
    {
      "entropy": 5.60113205909729,
      "epoch": 3.2170718432510883,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039822410616505545,
      "loss": 5.2462,
      "mean_token_accuracy": 0.18772169202566147,
      "num_tokens": 73585733.0,
      "step": 35465
    },
    {
      "entropy": 5.615761613845825,
      "epoch": 3.2175253991291726,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039819703230000006,
      "loss": 5.2389,
      "mean_token_accuracy": 0.19051095694303513,
      "num_tokens": 73597077.0,
      "step": 35470
    },
    {
      "entropy": 5.505122184753418,
      "epoch": 3.217978955007257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003981699558871986,
      "loss": 5.1958,
      "mean_token_accuracy": 0.1966714709997177,
      "num_tokens": 73607885.0,
      "step": 35475
    },
    {
      "entropy": 5.575867319107056,
      "epoch": 3.218432510885341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000398142876927211,
      "loss": 5.1701,
      "mean_token_accuracy": 0.19502515643835067,
      "num_tokens": 73618769.0,
      "step": 35480
    },
    {
      "entropy": 5.611887741088867,
      "epoch": 3.2188860667634254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039811579542059744,
      "loss": 5.25,
      "mean_token_accuracy": 0.18959474861621856,
      "num_tokens": 73629345.0,
      "step": 35485
    },
    {
      "entropy": 5.5403999328613285,
      "epoch": 3.2193396226415096,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039808871136791786,
      "loss": 5.1264,
      "mean_token_accuracy": 0.19559999406337739,
      "num_tokens": 73639698.0,
      "step": 35490
    },
    {
      "entropy": 5.498665285110474,
      "epoch": 3.2197931785195935,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039806162476973227,
      "loss": 5.1407,
      "mean_token_accuracy": 0.1954943284392357,
      "num_tokens": 73650954.0,
      "step": 35495
    },
    {
      "entropy": 5.5661036491394045,
      "epoch": 3.2202467343976777,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000398034535626601,
      "loss": 5.2597,
      "mean_token_accuracy": 0.18803949505090714,
      "num_tokens": 73661195.0,
      "step": 35500
    },
    {
      "entropy": 5.709169101715088,
      "epoch": 3.220700290275762,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039800744393908414,
      "loss": 5.2775,
      "mean_token_accuracy": 0.1894337862730026,
      "num_tokens": 73670512.0,
      "step": 35505
    },
    {
      "entropy": 5.594583415985108,
      "epoch": 3.2211538461538463,
      "grad_norm": 1.09375,
      "learning_rate": 0.000397980349707742,
      "loss": 5.2074,
      "mean_token_accuracy": 0.1887025535106659,
      "num_tokens": 73681927.0,
      "step": 35510
    },
    {
      "entropy": 5.572396230697632,
      "epoch": 3.2216074020319305,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039795325293313493,
      "loss": 5.2262,
      "mean_token_accuracy": 0.18944125026464462,
      "num_tokens": 73692025.0,
      "step": 35515
    },
    {
      "entropy": 5.504638624191284,
      "epoch": 3.2220609579100143,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039792615361582324,
      "loss": 5.1557,
      "mean_token_accuracy": 0.18920761942863465,
      "num_tokens": 73702278.0,
      "step": 35520
    },
    {
      "entropy": 5.655825281143189,
      "epoch": 3.2225145137880986,
      "grad_norm": 1.5,
      "learning_rate": 0.00039789905175636726,
      "loss": 5.2803,
      "mean_token_accuracy": 0.18197675794363022,
      "num_tokens": 73713511.0,
      "step": 35525
    },
    {
      "entropy": 5.498130655288696,
      "epoch": 3.222968069666183,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039787194735532765,
      "loss": 5.0624,
      "mean_token_accuracy": 0.20025645047426224,
      "num_tokens": 73723683.0,
      "step": 35530
    },
    {
      "entropy": 5.55171422958374,
      "epoch": 3.223421625544267,
      "grad_norm": 1.125,
      "learning_rate": 0.0003978448404132649,
      "loss": 5.1439,
      "mean_token_accuracy": 0.20186953842639924,
      "num_tokens": 73734191.0,
      "step": 35535
    },
    {
      "entropy": 5.565068864822388,
      "epoch": 3.2238751814223514,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039781773093073956,
      "loss": 5.1824,
      "mean_token_accuracy": 0.1894876942038536,
      "num_tokens": 73743747.0,
      "step": 35540
    },
    {
      "entropy": 5.5513139247894285,
      "epoch": 3.2243287373004352,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003977906189083122,
      "loss": 5.135,
      "mean_token_accuracy": 0.19749819934368135,
      "num_tokens": 73754466.0,
      "step": 35545
    },
    {
      "entropy": 5.569976806640625,
      "epoch": 3.2247822931785195,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0003977635043465435,
      "loss": 5.134,
      "mean_token_accuracy": 0.19535199999809266,
      "num_tokens": 73765948.0,
      "step": 35550
    },
    {
      "entropy": 5.4709193229675295,
      "epoch": 3.2252358490566038,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003977363872459943,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1959182873368263,
      "num_tokens": 73776003.0,
      "step": 35555
    },
    {
      "entropy": 5.620408582687378,
      "epoch": 3.225689404934688,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003977092676072254,
      "loss": 5.3382,
      "mean_token_accuracy": 0.19214208871126176,
      "num_tokens": 73787821.0,
      "step": 35560
    },
    {
      "entropy": 5.591610670089722,
      "epoch": 3.2261429608127723,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039768214543079746,
      "loss": 5.1775,
      "mean_token_accuracy": 0.18644013851881028,
      "num_tokens": 73799207.0,
      "step": 35565
    },
    {
      "entropy": 5.5356427192687985,
      "epoch": 3.226596516690856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003976550207172716,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1979799121618271,
      "num_tokens": 73809987.0,
      "step": 35570
    },
    {
      "entropy": 5.434296560287476,
      "epoch": 3.2270500725689404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039762789346720854,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19231218099594116,
      "num_tokens": 73819075.0,
      "step": 35575
    },
    {
      "entropy": 5.5660679817199705,
      "epoch": 3.2275036284470247,
      "grad_norm": 1.25,
      "learning_rate": 0.0003976007636811694,
      "loss": 5.2028,
      "mean_token_accuracy": 0.19776130616664886,
      "num_tokens": 73829831.0,
      "step": 35580
    },
    {
      "entropy": 5.568068122863769,
      "epoch": 3.227957184325109,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003975736313597153,
      "loss": 5.1727,
      "mean_token_accuracy": 0.19920382052659988,
      "num_tokens": 73839826.0,
      "step": 35585
    },
    {
      "entropy": 5.5555627822875975,
      "epoch": 3.228410740203193,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003975464965034072,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19760971814393996,
      "num_tokens": 73850006.0,
      "step": 35590
    },
    {
      "entropy": 5.578938674926758,
      "epoch": 3.228864296081277,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003975193591128064,
      "loss": 5.1435,
      "mean_token_accuracy": 0.19683133512735368,
      "num_tokens": 73860309.0,
      "step": 35595
    },
    {
      "entropy": 5.527131605148315,
      "epoch": 3.2293178519593613,
      "grad_norm": 2.359375,
      "learning_rate": 0.00039749221918847406,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18501399457454681,
      "num_tokens": 73871613.0,
      "step": 35600
    },
    {
      "entropy": 5.458112382888794,
      "epoch": 3.2297714078374455,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003974650767309713,
      "loss": 5.0768,
      "mean_token_accuracy": 0.19718154966831208,
      "num_tokens": 73882303.0,
      "step": 35605
    },
    {
      "entropy": 5.632185077667236,
      "epoch": 3.23022496371553,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039743793174085965,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18627649843692778,
      "num_tokens": 73893548.0,
      "step": 35610
    },
    {
      "entropy": 5.600238466262818,
      "epoch": 3.230678519593614,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00039741078421870024,
      "loss": 5.1337,
      "mean_token_accuracy": 0.1947312831878662,
      "num_tokens": 73903934.0,
      "step": 35615
    },
    {
      "entropy": 5.550078248977661,
      "epoch": 3.231132075471698,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00039738363416505474,
      "loss": 5.2502,
      "mean_token_accuracy": 0.18799962997436523,
      "num_tokens": 73913813.0,
      "step": 35620
    },
    {
      "entropy": 5.588428926467896,
      "epoch": 3.231585631349782,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039735648158048444,
      "loss": 5.1893,
      "mean_token_accuracy": 0.19446159452199935,
      "num_tokens": 73924339.0,
      "step": 35625
    },
    {
      "entropy": 5.55303692817688,
      "epoch": 3.2320391872278664,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003973293264655509,
      "loss": 5.1596,
      "mean_token_accuracy": 0.1867993637919426,
      "num_tokens": 73935318.0,
      "step": 35630
    },
    {
      "entropy": 5.551324081420899,
      "epoch": 3.2324927431059507,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003973021688208158,
      "loss": 5.1288,
      "mean_token_accuracy": 0.19952743053436278,
      "num_tokens": 73945547.0,
      "step": 35635
    },
    {
      "entropy": 5.53488712310791,
      "epoch": 3.232946298984035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003972750086468406,
      "loss": 5.1517,
      "mean_token_accuracy": 0.19315426349639891,
      "num_tokens": 73956195.0,
      "step": 35640
    },
    {
      "entropy": 5.5286163806915285,
      "epoch": 3.2333998548621192,
      "grad_norm": 1.6875,
      "learning_rate": 0.00039724784594418714,
      "loss": 5.1066,
      "mean_token_accuracy": 0.2009713664650917,
      "num_tokens": 73966843.0,
      "step": 35645
    },
    {
      "entropy": 5.566137742996216,
      "epoch": 3.233853410740203,
      "grad_norm": 1.0,
      "learning_rate": 0.000397220680713417,
      "loss": 5.1505,
      "mean_token_accuracy": 0.1979561284184456,
      "num_tokens": 73978516.0,
      "step": 35650
    },
    {
      "entropy": 5.4872723579406735,
      "epoch": 3.2343069666182873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039719351295509203,
      "loss": 5.0829,
      "mean_token_accuracy": 0.20257036983966828,
      "num_tokens": 73988376.0,
      "step": 35655
    },
    {
      "entropy": 5.533315515518188,
      "epoch": 3.2347605224963716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039716634266977413,
      "loss": 5.1218,
      "mean_token_accuracy": 0.20270778387784957,
      "num_tokens": 73998232.0,
      "step": 35660
    },
    {
      "entropy": 5.521911907196045,
      "epoch": 3.235214078374456,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003971391698580252,
      "loss": 5.125,
      "mean_token_accuracy": 0.1956624612212181,
      "num_tokens": 74008726.0,
      "step": 35665
    },
    {
      "entropy": 5.518981981277466,
      "epoch": 3.23566763425254,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039711199452040696,
      "loss": 5.1543,
      "mean_token_accuracy": 0.19573801904916763,
      "num_tokens": 74018746.0,
      "step": 35670
    },
    {
      "entropy": 5.508972072601319,
      "epoch": 3.236121190130624,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039708481665748157,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19824142009019852,
      "num_tokens": 74028615.0,
      "step": 35675
    },
    {
      "entropy": 5.532364416122436,
      "epoch": 3.236574746008708,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003970576362698111,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1917991206049919,
      "num_tokens": 74039219.0,
      "step": 35680
    },
    {
      "entropy": 5.54244031906128,
      "epoch": 3.2370283018867925,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00039703045335795766,
      "loss": 5.1369,
      "mean_token_accuracy": 0.19483816027641296,
      "num_tokens": 74049385.0,
      "step": 35685
    },
    {
      "entropy": 5.6046241283416744,
      "epoch": 3.2374818577648767,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003970032679224833,
      "loss": 5.301,
      "mean_token_accuracy": 0.17513615339994432,
      "num_tokens": 74059437.0,
      "step": 35690
    },
    {
      "entropy": 5.583999872207642,
      "epoch": 3.237935413642961,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039697607996395023,
      "loss": 5.2221,
      "mean_token_accuracy": 0.1918392777442932,
      "num_tokens": 74068834.0,
      "step": 35695
    },
    {
      "entropy": 5.592256546020508,
      "epoch": 3.238388969521045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003969488894829208,
      "loss": 5.1835,
      "mean_token_accuracy": 0.18867270946502684,
      "num_tokens": 74079757.0,
      "step": 35700
    },
    {
      "entropy": 5.5581520080566404,
      "epoch": 3.238842525399129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003969216964799573,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1952373668551445,
      "num_tokens": 74091092.0,
      "step": 35705
    },
    {
      "entropy": 5.486856603622437,
      "epoch": 3.2392960812772134,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003968945009556219,
      "loss": 5.1593,
      "mean_token_accuracy": 0.1946376383304596,
      "num_tokens": 74102272.0,
      "step": 35710
    },
    {
      "entropy": 5.50285382270813,
      "epoch": 3.2397496371552976,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003968673029104773,
      "loss": 5.1142,
      "mean_token_accuracy": 0.1998566895723343,
      "num_tokens": 74113101.0,
      "step": 35715
    },
    {
      "entropy": 5.612504863739014,
      "epoch": 3.240203193033382,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003968401023450858,
      "loss": 5.1399,
      "mean_token_accuracy": 0.20050008744001388,
      "num_tokens": 74123180.0,
      "step": 35720
    },
    {
      "entropy": 5.538495063781738,
      "epoch": 3.2406567489114657,
      "grad_norm": 1.125,
      "learning_rate": 0.00039681289926000985,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19330138713121414,
      "num_tokens": 74133675.0,
      "step": 35725
    },
    {
      "entropy": 5.451778411865234,
      "epoch": 3.24111030478955,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003967856936558122,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19784990400075914,
      "num_tokens": 74143472.0,
      "step": 35730
    },
    {
      "entropy": 5.539042377471924,
      "epoch": 3.2415638606676342,
      "grad_norm": 1.59375,
      "learning_rate": 0.00039675848553305526,
      "loss": 5.1234,
      "mean_token_accuracy": 0.1934482052922249,
      "num_tokens": 74154682.0,
      "step": 35735
    },
    {
      "entropy": 5.568213653564453,
      "epoch": 3.2420174165457185,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003967312748923019,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1926383778452873,
      "num_tokens": 74164745.0,
      "step": 35740
    },
    {
      "entropy": 5.46427354812622,
      "epoch": 3.2424709724238028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039670406173411473,
      "loss": 5.0806,
      "mean_token_accuracy": 0.19933598041534423,
      "num_tokens": 74175388.0,
      "step": 35745
    },
    {
      "entropy": 5.504526519775391,
      "epoch": 3.2429245283018866,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003966768460590566,
      "loss": 5.1138,
      "mean_token_accuracy": 0.1932731881737709,
      "num_tokens": 74185929.0,
      "step": 35750
    },
    {
      "entropy": 5.551115798950195,
      "epoch": 3.243378084179971,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039664962786769013,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19634269326925277,
      "num_tokens": 74196165.0,
      "step": 35755
    },
    {
      "entropy": 5.503727722167969,
      "epoch": 3.243831640058055,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039662240716057854,
      "loss": 5.1192,
      "mean_token_accuracy": 0.1984558656811714,
      "num_tokens": 74206843.0,
      "step": 35760
    },
    {
      "entropy": 5.508649301528931,
      "epoch": 3.2442851959361394,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003965951839382845,
      "loss": 5.121,
      "mean_token_accuracy": 0.19749027937650682,
      "num_tokens": 74216854.0,
      "step": 35765
    },
    {
      "entropy": 5.484331321716309,
      "epoch": 3.2447387518142237,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000396567958201371,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19392049610614776,
      "num_tokens": 74226158.0,
      "step": 35770
    },
    {
      "entropy": 5.539815855026245,
      "epoch": 3.2451923076923075,
      "grad_norm": 1.125,
      "learning_rate": 0.0003965407299504013,
      "loss": 5.1919,
      "mean_token_accuracy": 0.19537965059280396,
      "num_tokens": 74236860.0,
      "step": 35775
    },
    {
      "entropy": 5.581633901596069,
      "epoch": 3.2456458635703918,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0003965134991859382,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19602890759706498,
      "num_tokens": 74247706.0,
      "step": 35780
    },
    {
      "entropy": 5.580098104476929,
      "epoch": 3.246099419448476,
      "grad_norm": 1.234375,
      "learning_rate": 0.000396486265908545,
      "loss": 5.194,
      "mean_token_accuracy": 0.1941370889544487,
      "num_tokens": 74259479.0,
      "step": 35785
    },
    {
      "entropy": 5.497936344146728,
      "epoch": 3.2465529753265603,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003964590301187849,
      "loss": 5.1931,
      "mean_token_accuracy": 0.1964995250105858,
      "num_tokens": 74270371.0,
      "step": 35790
    },
    {
      "entropy": 5.556675624847412,
      "epoch": 3.2470065312046446,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003964317918172211,
      "loss": 5.1136,
      "mean_token_accuracy": 0.20864863097667694,
      "num_tokens": 74279637.0,
      "step": 35795
    },
    {
      "entropy": 5.55341854095459,
      "epoch": 3.247460087082729,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00039640455100441684,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1902659460902214,
      "num_tokens": 74289889.0,
      "step": 35800
    },
    {
      "entropy": 5.573930835723877,
      "epoch": 3.2479136429608126,
      "grad_norm": 2.125,
      "learning_rate": 0.00039637730768093564,
      "loss": 5.1863,
      "mean_token_accuracy": 0.19926695972681047,
      "num_tokens": 74300949.0,
      "step": 35805
    },
    {
      "entropy": 5.571553325653076,
      "epoch": 3.248367198838897,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003963500618473407,
      "loss": 5.1144,
      "mean_token_accuracy": 0.19732340276241303,
      "num_tokens": 74312077.0,
      "step": 35810
    },
    {
      "entropy": 5.506304836273193,
      "epoch": 3.248820754716981,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003963228135041956,
      "loss": 5.1867,
      "mean_token_accuracy": 0.19206663221120834,
      "num_tokens": 74322610.0,
      "step": 35815
    },
    {
      "entropy": 5.454351711273193,
      "epoch": 3.2492743105950654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039629556265206373,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20162488371133805,
      "num_tokens": 74331970.0,
      "step": 35820
    },
    {
      "entropy": 5.51569619178772,
      "epoch": 3.2497278664731493,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003962683092915088,
      "loss": 5.1039,
      "mean_token_accuracy": 0.1981153205037117,
      "num_tokens": 74342812.0,
      "step": 35825
    },
    {
      "entropy": 5.579697132110596,
      "epoch": 3.2501814223512335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003962410534230943,
      "loss": 5.0807,
      "mean_token_accuracy": 0.1992013245820999,
      "num_tokens": 74353786.0,
      "step": 35830
    },
    {
      "entropy": 5.536157894134521,
      "epoch": 3.250634978229318,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039621379504738385,
      "loss": 5.1829,
      "mean_token_accuracy": 0.19850765019655228,
      "num_tokens": 74363732.0,
      "step": 35835
    },
    {
      "entropy": 5.523733520507813,
      "epoch": 3.251088534107402,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039618653416494137,
      "loss": 5.1848,
      "mean_token_accuracy": 0.19452111721038817,
      "num_tokens": 74374077.0,
      "step": 35840
    },
    {
      "entropy": 5.579748678207397,
      "epoch": 3.2515420899854863,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039615927077633043,
      "loss": 5.1922,
      "mean_token_accuracy": 0.1939806416630745,
      "num_tokens": 74384758.0,
      "step": 35845
    },
    {
      "entropy": 5.668642854690551,
      "epoch": 3.2519956458635706,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039613200488211483,
      "loss": 5.2259,
      "mean_token_accuracy": 0.19402261227369308,
      "num_tokens": 74395811.0,
      "step": 35850
    },
    {
      "entropy": 5.534943675994873,
      "epoch": 3.2524492017416544,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00039610473648285857,
      "loss": 5.2005,
      "mean_token_accuracy": 0.19440939128398896,
      "num_tokens": 74407830.0,
      "step": 35855
    },
    {
      "entropy": 5.4671083927154545,
      "epoch": 3.2529027576197387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003960774655791255,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19973210841417313,
      "num_tokens": 74417565.0,
      "step": 35860
    },
    {
      "entropy": 5.56940598487854,
      "epoch": 3.253356313497823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039605019217147947,
      "loss": 5.1666,
      "mean_token_accuracy": 0.1918021097779274,
      "num_tokens": 74428015.0,
      "step": 35865
    },
    {
      "entropy": 5.492491006851196,
      "epoch": 3.253809869375907,
      "grad_norm": 5.21875,
      "learning_rate": 0.00039602291626048475,
      "loss": 5.1641,
      "mean_token_accuracy": 0.2008276477456093,
      "num_tokens": 74439523.0,
      "step": 35870
    },
    {
      "entropy": 5.604525709152222,
      "epoch": 3.2542634252539915,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00039599563784670526,
      "loss": 5.229,
      "mean_token_accuracy": 0.18606900423765182,
      "num_tokens": 74449897.0,
      "step": 35875
    },
    {
      "entropy": 5.585801601409912,
      "epoch": 3.2547169811320753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003959683569307051,
      "loss": 5.0714,
      "mean_token_accuracy": 0.20492848008871078,
      "num_tokens": 74459888.0,
      "step": 35880
    },
    {
      "entropy": 5.496631288528443,
      "epoch": 3.2551705370101596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003959410735130485,
      "loss": 5.0281,
      "mean_token_accuracy": 0.19902896136045456,
      "num_tokens": 74470894.0,
      "step": 35885
    },
    {
      "entropy": 5.4623480319976805,
      "epoch": 3.255624092888244,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039591378759429974,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20291269272565843,
      "num_tokens": 74481306.0,
      "step": 35890
    },
    {
      "entropy": 5.493529796600342,
      "epoch": 3.256077648766328,
      "grad_norm": 1.390625,
      "learning_rate": 0.000395886499175023,
      "loss": 5.1364,
      "mean_token_accuracy": 0.19281182140111924,
      "num_tokens": 74491048.0,
      "step": 35895
    },
    {
      "entropy": 5.500912046432495,
      "epoch": 3.2565312046444124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039585920825578266,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20306481420993805,
      "num_tokens": 74501656.0,
      "step": 35900
    },
    {
      "entropy": 5.575534248352051,
      "epoch": 3.256984760522496,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003958319148371431,
      "loss": 5.1982,
      "mean_token_accuracy": 0.19243727773427963,
      "num_tokens": 74513280.0,
      "step": 35905
    },
    {
      "entropy": 5.537912464141845,
      "epoch": 3.2574383164005805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003958046189196687,
      "loss": 5.18,
      "mean_token_accuracy": 0.19611378759145737,
      "num_tokens": 74522898.0,
      "step": 35910
    },
    {
      "entropy": 5.645910167694092,
      "epoch": 3.2578918722786647,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039577732050392395,
      "loss": 5.2046,
      "mean_token_accuracy": 0.1926321119070053,
      "num_tokens": 74532551.0,
      "step": 35915
    },
    {
      "entropy": 5.526605796813965,
      "epoch": 3.258345428156749,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003957500195904735,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2029591366648674,
      "num_tokens": 74542285.0,
      "step": 35920
    },
    {
      "entropy": 5.51683611869812,
      "epoch": 3.2587989840348333,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000395722716179882,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1978670120239258,
      "num_tokens": 74551982.0,
      "step": 35925
    },
    {
      "entropy": 5.5496618270874025,
      "epoch": 3.259252539912917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003956954102727138,
      "loss": 5.1684,
      "mean_token_accuracy": 0.19248173087835313,
      "num_tokens": 74562190.0,
      "step": 35930
    },
    {
      "entropy": 5.496221351623535,
      "epoch": 3.2597060957910013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039566810186953376,
      "loss": 5.0714,
      "mean_token_accuracy": 0.19912100881338118,
      "num_tokens": 74572673.0,
      "step": 35935
    },
    {
      "entropy": 5.501903295516968,
      "epoch": 3.2601596516690856,
      "grad_norm": 1.125,
      "learning_rate": 0.0003956407909709068,
      "loss": 5.0778,
      "mean_token_accuracy": 0.20320799946784973,
      "num_tokens": 74582686.0,
      "step": 35940
    },
    {
      "entropy": 5.607358837127686,
      "epoch": 3.26061320754717,
      "grad_norm": 1.125,
      "learning_rate": 0.00039561347757739734,
      "loss": 5.235,
      "mean_token_accuracy": 0.19530455768108368,
      "num_tokens": 74593658.0,
      "step": 35945
    },
    {
      "entropy": 5.627692556381225,
      "epoch": 3.261066763425254,
      "grad_norm": 2.015625,
      "learning_rate": 0.00039558616168957044,
      "loss": 5.2418,
      "mean_token_accuracy": 0.1938393995165825,
      "num_tokens": 74603585.0,
      "step": 35950
    },
    {
      "entropy": 5.519414854049683,
      "epoch": 3.2615203193033384,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003955588433079911,
      "loss": 5.1025,
      "mean_token_accuracy": 0.2028382107615471,
      "num_tokens": 74612806.0,
      "step": 35955
    },
    {
      "entropy": 5.558960103988648,
      "epoch": 3.2619738751814222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039553152243322414,
      "loss": 5.1883,
      "mean_token_accuracy": 0.19970777332782746,
      "num_tokens": 74623252.0,
      "step": 35960
    },
    {
      "entropy": 5.549375391006469,
      "epoch": 3.2624274310595065,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003955041990658346,
      "loss": 5.1279,
      "mean_token_accuracy": 0.19918239265680313,
      "num_tokens": 74633555.0,
      "step": 35965
    },
    {
      "entropy": 5.525870275497437,
      "epoch": 3.2628809869375908,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003954768732063874,
      "loss": 5.1317,
      "mean_token_accuracy": 0.19993505179882048,
      "num_tokens": 74643322.0,
      "step": 35970
    },
    {
      "entropy": 5.48020830154419,
      "epoch": 3.263334542815675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003954495448554478,
      "loss": 5.1156,
      "mean_token_accuracy": 0.20231548994779586,
      "num_tokens": 74653461.0,
      "step": 35975
    },
    {
      "entropy": 5.511305475234986,
      "epoch": 3.263788098693759,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039542221401358104,
      "loss": 5.1861,
      "mean_token_accuracy": 0.19254685640335084,
      "num_tokens": 74663470.0,
      "step": 35980
    },
    {
      "entropy": 5.538217830657959,
      "epoch": 3.264241654571843,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000395394880681352,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20423410832881927,
      "num_tokens": 74674074.0,
      "step": 35985
    },
    {
      "entropy": 5.606934118270874,
      "epoch": 3.2646952104499274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039536754485932626,
      "loss": 5.254,
      "mean_token_accuracy": 0.18563302457332612,
      "num_tokens": 74684968.0,
      "step": 35990
    },
    {
      "entropy": 5.492776870727539,
      "epoch": 3.2651487663280117,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039534020654806906,
      "loss": 5.0754,
      "mean_token_accuracy": 0.20745983421802522,
      "num_tokens": 74694776.0,
      "step": 35995
    },
    {
      "entropy": 5.5084065914154055,
      "epoch": 3.265602322206096,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039531286574814563,
      "loss": 5.0984,
      "mean_token_accuracy": 0.1963493138551712,
      "num_tokens": 74705506.0,
      "step": 36000
    },
    {
      "epoch": 3.265602322206096,
      "eval_entropy": 5.290731865787937,
      "eval_loss": 5.231833457946777,
      "eval_mean_token_accuracy": 0.20003910623322246,
      "eval_num_tokens": 74705506.0,
      "eval_runtime": 26.7883,
      "eval_samples_per_second": 1319.978,
      "eval_steps_per_second": 164.997,
      "step": 36000
    },
    {
      "entropy": 5.502193737030029,
      "epoch": 3.26605587808418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003952855224601215,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19676531702280045,
      "num_tokens": 74715677.0,
      "step": 36005
    },
    {
      "entropy": 5.571337699890137,
      "epoch": 3.266509433962264,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003952581766845622,
      "loss": 5.1159,
      "mean_token_accuracy": 0.20332637578248977,
      "num_tokens": 74725296.0,
      "step": 36010
    },
    {
      "entropy": 5.479864645004272,
      "epoch": 3.2669629898403483,
      "grad_norm": 1.375,
      "learning_rate": 0.0003952308284220331,
      "loss": 5.1026,
      "mean_token_accuracy": 0.20366611033678056,
      "num_tokens": 74736124.0,
      "step": 36015
    },
    {
      "entropy": 5.3931300163269045,
      "epoch": 3.2674165457184325,
      "grad_norm": 2.046875,
      "learning_rate": 0.00039520347767309975,
      "loss": 5.1539,
      "mean_token_accuracy": 0.18986068218946456,
      "num_tokens": 74746447.0,
      "step": 36020
    },
    {
      "entropy": 5.560318326950073,
      "epoch": 3.267870101596517,
      "grad_norm": 1.046875,
      "learning_rate": 0.000395176124438328,
      "loss": 5.1517,
      "mean_token_accuracy": 0.19088142812252046,
      "num_tokens": 74756947.0,
      "step": 36025
    },
    {
      "entropy": 5.542709445953369,
      "epoch": 3.2683236574746006,
      "grad_norm": 2.421875,
      "learning_rate": 0.00039514876871828326,
      "loss": 5.07,
      "mean_token_accuracy": 0.2023648053407669,
      "num_tokens": 74767856.0,
      "step": 36030
    },
    {
      "entropy": 5.491515016555786,
      "epoch": 3.268777213352685,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039512141051353144,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19717976152896882,
      "num_tokens": 74778335.0,
      "step": 36035
    },
    {
      "entropy": 5.544978237152099,
      "epoch": 3.269230769230769,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003950940498246382,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19784400463104249,
      "num_tokens": 74788430.0,
      "step": 36040
    },
    {
      "entropy": 5.618734884262085,
      "epoch": 3.2696843251088534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039506668665216944,
      "loss": 5.2561,
      "mean_token_accuracy": 0.1869785815477371,
      "num_tokens": 74800193.0,
      "step": 36045
    },
    {
      "entropy": 5.56521897315979,
      "epoch": 3.2701378809869377,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000395039320996691,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18907537162303925,
      "num_tokens": 74810368.0,
      "step": 36050
    },
    {
      "entropy": 5.51762957572937,
      "epoch": 3.270591436865022,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039501195285876874,
      "loss": 5.1264,
      "mean_token_accuracy": 0.20162811130285263,
      "num_tokens": 74820687.0,
      "step": 36055
    },
    {
      "entropy": 5.4735301494598385,
      "epoch": 3.271044992743106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003949845822389688,
      "loss": 5.0878,
      "mean_token_accuracy": 0.19746065139770508,
      "num_tokens": 74830489.0,
      "step": 36060
    },
    {
      "entropy": 5.426382207870484,
      "epoch": 3.27149854862119,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000394957209137857,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19967115074396133,
      "num_tokens": 74840638.0,
      "step": 36065
    },
    {
      "entropy": 5.600813388824463,
      "epoch": 3.2719521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003949298335559997,
      "loss": 5.2785,
      "mean_token_accuracy": 0.18449120670557023,
      "num_tokens": 74850600.0,
      "step": 36070
    },
    {
      "entropy": 5.586358594894409,
      "epoch": 3.2724056603773586,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003949024554939628,
      "loss": 5.1609,
      "mean_token_accuracy": 0.1981416329741478,
      "num_tokens": 74861177.0,
      "step": 36075
    },
    {
      "entropy": 5.560961484909058,
      "epoch": 3.272859216255443,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039487507495231263,
      "loss": 5.2216,
      "mean_token_accuracy": 0.1927141010761261,
      "num_tokens": 74871747.0,
      "step": 36080
    },
    {
      "entropy": 5.5221773147583,
      "epoch": 3.2733127721335267,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003948476919316153,
      "loss": 5.1595,
      "mean_token_accuracy": 0.19554524570703508,
      "num_tokens": 74882796.0,
      "step": 36085
    },
    {
      "entropy": 5.562075424194336,
      "epoch": 3.273766328011611,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003948203064324371,
      "loss": 5.133,
      "mean_token_accuracy": 0.19789946526288987,
      "num_tokens": 74894123.0,
      "step": 36090
    },
    {
      "entropy": 5.545917129516601,
      "epoch": 3.274219883889695,
      "grad_norm": 2.578125,
      "learning_rate": 0.00039479291845534453,
      "loss": 5.1838,
      "mean_token_accuracy": 0.18825606852769852,
      "num_tokens": 74904787.0,
      "step": 36095
    },
    {
      "entropy": 5.586173248291016,
      "epoch": 3.2746734397677795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039476552800090384,
      "loss": 5.2117,
      "mean_token_accuracy": 0.1865643620491028,
      "num_tokens": 74915041.0,
      "step": 36100
    },
    {
      "entropy": 5.584884166717529,
      "epoch": 3.2751269956458637,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039473813506968146,
      "loss": 5.1195,
      "mean_token_accuracy": 0.1912941336631775,
      "num_tokens": 74926217.0,
      "step": 36105
    },
    {
      "entropy": 5.618125057220459,
      "epoch": 3.2755805515239476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039471073966224404,
      "loss": 5.2046,
      "mean_token_accuracy": 0.1905893102288246,
      "num_tokens": 74936616.0,
      "step": 36110
    },
    {
      "entropy": 5.579011535644531,
      "epoch": 3.276034107402032,
      "grad_norm": 2.09375,
      "learning_rate": 0.000394683341779158,
      "loss": 5.2033,
      "mean_token_accuracy": 0.19057361632585526,
      "num_tokens": 74947443.0,
      "step": 36115
    },
    {
      "entropy": 5.521907234191895,
      "epoch": 3.276487663280116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003946559414209899,
      "loss": 5.1004,
      "mean_token_accuracy": 0.20001330375671386,
      "num_tokens": 74957618.0,
      "step": 36120
    },
    {
      "entropy": 5.461640310287476,
      "epoch": 3.2769412191582004,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003946285385883065,
      "loss": 5.0658,
      "mean_token_accuracy": 0.1981511503458023,
      "num_tokens": 74967982.0,
      "step": 36125
    },
    {
      "entropy": 5.497540473937988,
      "epoch": 3.2773947750362846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039460113328167434,
      "loss": 5.1425,
      "mean_token_accuracy": 0.1998114302754402,
      "num_tokens": 74978801.0,
      "step": 36130
    },
    {
      "entropy": 5.521474981307984,
      "epoch": 3.2778483309143684,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003945737255016603,
      "loss": 5.0793,
      "mean_token_accuracy": 0.19852500259876252,
      "num_tokens": 74989000.0,
      "step": 36135
    },
    {
      "entropy": 5.521269702911377,
      "epoch": 3.2783018867924527,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039454631524883123,
      "loss": 5.1961,
      "mean_token_accuracy": 0.1938331753015518,
      "num_tokens": 74999664.0,
      "step": 36140
    },
    {
      "entropy": 5.530513286590576,
      "epoch": 3.278755442670537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003945189025237538,
      "loss": 5.1689,
      "mean_token_accuracy": 0.19818912744522094,
      "num_tokens": 75010187.0,
      "step": 36145
    },
    {
      "entropy": 5.66373028755188,
      "epoch": 3.2792089985486212,
      "grad_norm": 1.015625,
      "learning_rate": 0.000394491487326995,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1879044622182846,
      "num_tokens": 75021222.0,
      "step": 36150
    },
    {
      "entropy": 5.509217357635498,
      "epoch": 3.2796625544267055,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003944640696591219,
      "loss": 5.1587,
      "mean_token_accuracy": 0.20102105140686036,
      "num_tokens": 75031136.0,
      "step": 36155
    },
    {
      "entropy": 5.4269033908844,
      "epoch": 3.28011611030479,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039443664952070135,
      "loss": 5.1084,
      "mean_token_accuracy": 0.2015734851360321,
      "num_tokens": 75041545.0,
      "step": 36160
    },
    {
      "entropy": 5.498659658432007,
      "epoch": 3.2805696661828736,
      "grad_norm": 1.0,
      "learning_rate": 0.00039440922691230044,
      "loss": 5.151,
      "mean_token_accuracy": 0.20327685922384262,
      "num_tokens": 75051805.0,
      "step": 36165
    },
    {
      "entropy": 5.536620044708252,
      "epoch": 3.281023222060958,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039438180183448625,
      "loss": 5.1406,
      "mean_token_accuracy": 0.20028531402349473,
      "num_tokens": 75062551.0,
      "step": 36170
    },
    {
      "entropy": 5.557577991485596,
      "epoch": 3.281476777939042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000394354374287826,
      "loss": 5.1377,
      "mean_token_accuracy": 0.19792572259902955,
      "num_tokens": 75073040.0,
      "step": 36175
    },
    {
      "entropy": 5.554858779907226,
      "epoch": 3.2819303338171264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039432694427288695,
      "loss": 5.119,
      "mean_token_accuracy": 0.20194688439369202,
      "num_tokens": 75082543.0,
      "step": 36180
    },
    {
      "entropy": 5.5397789001464846,
      "epoch": 3.28238388969521,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039429951179023614,
      "loss": 5.181,
      "mean_token_accuracy": 0.19515479654073714,
      "num_tokens": 75092480.0,
      "step": 36185
    },
    {
      "entropy": 5.486477947235107,
      "epoch": 3.2828374455732945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039427207684044116,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19803606420755387,
      "num_tokens": 75102445.0,
      "step": 36190
    },
    {
      "entropy": 5.568092727661133,
      "epoch": 3.2832910014513788,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039424463942406917,
      "loss": 5.2011,
      "mean_token_accuracy": 0.19205943793058394,
      "num_tokens": 75112793.0,
      "step": 36195
    },
    {
      "entropy": 5.598970508575439,
      "epoch": 3.283744557329463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003942171995416877,
      "loss": 5.1993,
      "mean_token_accuracy": 0.1945283070206642,
      "num_tokens": 75123312.0,
      "step": 36200
    },
    {
      "entropy": 5.479013824462891,
      "epoch": 3.2841981132075473,
      "grad_norm": 1.0,
      "learning_rate": 0.00039418975719386406,
      "loss": 5.0821,
      "mean_token_accuracy": 0.1983681544661522,
      "num_tokens": 75133584.0,
      "step": 36205
    },
    {
      "entropy": 5.530718994140625,
      "epoch": 3.2846516690856316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003941623123811659,
      "loss": 5.129,
      "mean_token_accuracy": 0.19664097279310228,
      "num_tokens": 75144344.0,
      "step": 36210
    },
    {
      "entropy": 5.588427972793579,
      "epoch": 3.2851052249637154,
      "grad_norm": 1.46875,
      "learning_rate": 0.00039413486510416087,
      "loss": 5.165,
      "mean_token_accuracy": 0.19132366627454758,
      "num_tokens": 75155086.0,
      "step": 36215
    },
    {
      "entropy": 5.517465114593506,
      "epoch": 3.2855587808417996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039410741536341634,
      "loss": 5.1053,
      "mean_token_accuracy": 0.1995728611946106,
      "num_tokens": 75166046.0,
      "step": 36220
    },
    {
      "entropy": 5.488533210754395,
      "epoch": 3.286012336719884,
      "grad_norm": 1.890625,
      "learning_rate": 0.00039407996315950016,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20072855651378632,
      "num_tokens": 75175388.0,
      "step": 36225
    },
    {
      "entropy": 5.5692400455474855,
      "epoch": 3.286465892597968,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039405250849297997,
      "loss": 5.1839,
      "mean_token_accuracy": 0.19852672517299652,
      "num_tokens": 75186213.0,
      "step": 36230
    },
    {
      "entropy": 5.490882062911988,
      "epoch": 3.2869194484760524,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003940250513644236,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19929418116807937,
      "num_tokens": 75196141.0,
      "step": 36235
    },
    {
      "entropy": 5.396495580673218,
      "epoch": 3.2873730043541363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003939975917743988,
      "loss": 4.9713,
      "mean_token_accuracy": 0.21407524347305298,
      "num_tokens": 75205414.0,
      "step": 36240
    },
    {
      "entropy": 5.549437141418457,
      "epoch": 3.2878265602322205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003939701297234735,
      "loss": 5.2011,
      "mean_token_accuracy": 0.19329699128866196,
      "num_tokens": 75215860.0,
      "step": 36245
    },
    {
      "entropy": 5.560226678848267,
      "epoch": 3.288280116110305,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039394266521221565,
      "loss": 5.157,
      "mean_token_accuracy": 0.1934611052274704,
      "num_tokens": 75226358.0,
      "step": 36250
    },
    {
      "entropy": 5.45194411277771,
      "epoch": 3.288733671988389,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039391519824119315,
      "loss": 4.9984,
      "mean_token_accuracy": 0.21000817120075227,
      "num_tokens": 75235779.0,
      "step": 36255
    },
    {
      "entropy": 5.567277622222901,
      "epoch": 3.2891872278664733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039388772881097396,
      "loss": 5.2309,
      "mean_token_accuracy": 0.19401153028011323,
      "num_tokens": 75246417.0,
      "step": 36260
    },
    {
      "entropy": 5.553352546691895,
      "epoch": 3.289640783744557,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039386025692212637,
      "loss": 5.1268,
      "mean_token_accuracy": 0.19879698753356934,
      "num_tokens": 75257094.0,
      "step": 36265
    },
    {
      "entropy": 5.455335855484009,
      "epoch": 3.2900943396226414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003938327825752183,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19696102142333985,
      "num_tokens": 75267374.0,
      "step": 36270
    },
    {
      "entropy": 5.463714742660523,
      "epoch": 3.2905478955007257,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000393805305770818,
      "loss": 5.1213,
      "mean_token_accuracy": 0.2061556488275528,
      "num_tokens": 75276936.0,
      "step": 36275
    },
    {
      "entropy": 5.503672981262207,
      "epoch": 3.29100145137881,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003937778265094937,
      "loss": 5.0678,
      "mean_token_accuracy": 0.20411434173583984,
      "num_tokens": 75286403.0,
      "step": 36280
    },
    {
      "entropy": 5.540531063079834,
      "epoch": 3.291455007256894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003937503447918137,
      "loss": 5.1659,
      "mean_token_accuracy": 0.19378619492053986,
      "num_tokens": 75297421.0,
      "step": 36285
    },
    {
      "entropy": 5.531997776031494,
      "epoch": 3.291908563134978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039372286061834633,
      "loss": 5.1665,
      "mean_token_accuracy": 0.19133200645446777,
      "num_tokens": 75307955.0,
      "step": 36290
    },
    {
      "entropy": 5.598216867446899,
      "epoch": 3.2923621190130623,
      "grad_norm": 1.546875,
      "learning_rate": 0.00039369537398966,
      "loss": 5.1866,
      "mean_token_accuracy": 0.1963188260793686,
      "num_tokens": 75318739.0,
      "step": 36295
    },
    {
      "entropy": 5.522044992446899,
      "epoch": 3.2928156748911466,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039366788490632286,
      "loss": 5.0158,
      "mean_token_accuracy": 0.20445054173469543,
      "num_tokens": 75328629.0,
      "step": 36300
    },
    {
      "entropy": 5.438244485855103,
      "epoch": 3.293269230769231,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039364039336890386,
      "loss": 5.0829,
      "mean_token_accuracy": 0.20429932624101638,
      "num_tokens": 75337986.0,
      "step": 36305
    },
    {
      "entropy": 5.4692795753479,
      "epoch": 3.293722786647315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003936128993779712,
      "loss": 5.0476,
      "mean_token_accuracy": 0.2012970358133316,
      "num_tokens": 75346169.0,
      "step": 36310
    },
    {
      "entropy": 5.561616277694702,
      "epoch": 3.2941763425253994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003935854029340936,
      "loss": 5.1509,
      "mean_token_accuracy": 0.20029715746641158,
      "num_tokens": 75356111.0,
      "step": 36315
    },
    {
      "entropy": 5.516036748886108,
      "epoch": 3.294629898403483,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039355790403783954,
      "loss": 5.0879,
      "mean_token_accuracy": 0.20116915553808212,
      "num_tokens": 75366679.0,
      "step": 36320
    },
    {
      "entropy": 5.452176952362061,
      "epoch": 3.2950834542815675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039353040268977785,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20870444625616075,
      "num_tokens": 75377500.0,
      "step": 36325
    },
    {
      "entropy": 5.483452081680298,
      "epoch": 3.2955370101596517,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003935028988904773,
      "loss": 5.101,
      "mean_token_accuracy": 0.1994316279888153,
      "num_tokens": 75387115.0,
      "step": 36330
    },
    {
      "entropy": 5.496873950958252,
      "epoch": 3.295990566037736,
      "grad_norm": 1.125,
      "learning_rate": 0.00039347539264050654,
      "loss": 5.1997,
      "mean_token_accuracy": 0.1913861081004143,
      "num_tokens": 75396330.0,
      "step": 36335
    },
    {
      "entropy": 5.516619729995727,
      "epoch": 3.29644412191582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003934478839404345,
      "loss": 5.1525,
      "mean_token_accuracy": 0.19937397688627242,
      "num_tokens": 75407168.0,
      "step": 36340
    },
    {
      "entropy": 5.6056074619293215,
      "epoch": 3.296897677793904,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039342037279083006,
      "loss": 5.178,
      "mean_token_accuracy": 0.19064828157424926,
      "num_tokens": 75418117.0,
      "step": 36345
    },
    {
      "entropy": 5.625768852233887,
      "epoch": 3.2973512336719883,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003933928591922621,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18896213918924332,
      "num_tokens": 75429026.0,
      "step": 36350
    },
    {
      "entropy": 5.538843870162964,
      "epoch": 3.2978047895500726,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003933653431452996,
      "loss": 5.1029,
      "mean_token_accuracy": 0.19750140309333802,
      "num_tokens": 75439710.0,
      "step": 36355
    },
    {
      "entropy": 5.485044384002686,
      "epoch": 3.298258345428157,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039333782465051173,
      "loss": 5.0979,
      "mean_token_accuracy": 0.20414044857025146,
      "num_tokens": 75450086.0,
      "step": 36360
    },
    {
      "entropy": 5.492785835266114,
      "epoch": 3.298711901306241,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003933103037084674,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19312057942152022,
      "num_tokens": 75460988.0,
      "step": 36365
    },
    {
      "entropy": 5.603398847579956,
      "epoch": 3.299165457184325,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00039328278031973586,
      "loss": 5.1905,
      "mean_token_accuracy": 0.19236952513456346,
      "num_tokens": 75471560.0,
      "step": 36370
    },
    {
      "entropy": 5.576504802703857,
      "epoch": 3.2996190130624092,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003932552544848863,
      "loss": 5.171,
      "mean_token_accuracy": 0.19297209978103638,
      "num_tokens": 75482124.0,
      "step": 36375
    },
    {
      "entropy": 5.518777465820312,
      "epoch": 3.3000725689404935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000393227726204488,
      "loss": 5.1305,
      "mean_token_accuracy": 0.20093966871500016,
      "num_tokens": 75492351.0,
      "step": 36380
    },
    {
      "entropy": 5.536232662200928,
      "epoch": 3.3005261248185778,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039320019547911013,
      "loss": 5.0637,
      "mean_token_accuracy": 0.2025340124964714,
      "num_tokens": 75502475.0,
      "step": 36385
    },
    {
      "entropy": 5.456142330169678,
      "epoch": 3.3009796806966616,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003931726623093221,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20571159422397614,
      "num_tokens": 75512345.0,
      "step": 36390
    },
    {
      "entropy": 5.459663724899292,
      "epoch": 3.301433236574746,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039314512669569323,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20964923799037932,
      "num_tokens": 75523011.0,
      "step": 36395
    },
    {
      "entropy": 5.530296897888183,
      "epoch": 3.30188679245283,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039311758863879316,
      "loss": 5.096,
      "mean_token_accuracy": 0.19810779988765717,
      "num_tokens": 75533775.0,
      "step": 36400
    },
    {
      "entropy": 5.534063291549683,
      "epoch": 3.3023403483309144,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039309004813919117,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18903688043355943,
      "num_tokens": 75544558.0,
      "step": 36405
    },
    {
      "entropy": 5.617114734649658,
      "epoch": 3.3027939042089987,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039306250519745696,
      "loss": 5.248,
      "mean_token_accuracy": 0.18658885955810547,
      "num_tokens": 75555599.0,
      "step": 36410
    },
    {
      "entropy": 5.525664901733398,
      "epoch": 3.303247460087083,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039303495981416007,
      "loss": 5.1125,
      "mean_token_accuracy": 0.20018726140260695,
      "num_tokens": 75566320.0,
      "step": 36415
    },
    {
      "entropy": 5.523219203948974,
      "epoch": 3.3037010159651667,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003930074119898701,
      "loss": 5.1708,
      "mean_token_accuracy": 0.19703798294067382,
      "num_tokens": 75576069.0,
      "step": 36420
    },
    {
      "entropy": 5.526902151107788,
      "epoch": 3.304154571843251,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003929798617251567,
      "loss": 5.1298,
      "mean_token_accuracy": 0.19895460307598115,
      "num_tokens": 75586012.0,
      "step": 36425
    },
    {
      "entropy": 5.56046986579895,
      "epoch": 3.3046081277213353,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003929523090205898,
      "loss": 5.2053,
      "mean_token_accuracy": 0.1915962040424347,
      "num_tokens": 75596499.0,
      "step": 36430
    },
    {
      "entropy": 5.509277820587158,
      "epoch": 3.3050616835994195,
      "grad_norm": 1.109375,
      "learning_rate": 0.000392924753876739,
      "loss": 5.0757,
      "mean_token_accuracy": 0.1983737513422966,
      "num_tokens": 75606329.0,
      "step": 36435
    },
    {
      "entropy": 5.53184404373169,
      "epoch": 3.305515239477504,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003928971962941743,
      "loss": 5.1286,
      "mean_token_accuracy": 0.1942514806985855,
      "num_tokens": 75617371.0,
      "step": 36440
    },
    {
      "entropy": 5.525479984283447,
      "epoch": 3.3059687953555876,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003928696362734655,
      "loss": 5.1222,
      "mean_token_accuracy": 0.1966325506567955,
      "num_tokens": 75627949.0,
      "step": 36445
    },
    {
      "entropy": 5.505697011947632,
      "epoch": 3.306422351233672,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039284207381518266,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1996284082531929,
      "num_tokens": 75638642.0,
      "step": 36450
    },
    {
      "entropy": 5.5062707424163815,
      "epoch": 3.306875907111756,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039281450891989555,
      "loss": 5.1361,
      "mean_token_accuracy": 0.19992248415946962,
      "num_tokens": 75649997.0,
      "step": 36455
    },
    {
      "entropy": 5.5241179943084715,
      "epoch": 3.3073294629898404,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003927869415881745,
      "loss": 5.1036,
      "mean_token_accuracy": 0.20336066335439681,
      "num_tokens": 75660223.0,
      "step": 36460
    },
    {
      "entropy": 5.526454019546509,
      "epoch": 3.3077830188679247,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003927593718205895,
      "loss": 5.0751,
      "mean_token_accuracy": 0.20387763381004334,
      "num_tokens": 75670550.0,
      "step": 36465
    },
    {
      "entropy": 5.53609881401062,
      "epoch": 3.3082365747460085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039273179961771065,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18628667145967484,
      "num_tokens": 75681631.0,
      "step": 36470
    },
    {
      "entropy": 5.499224662780762,
      "epoch": 3.308690130624093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003927042249801082,
      "loss": 5.0753,
      "mean_token_accuracy": 0.20629607737064362,
      "num_tokens": 75691561.0,
      "step": 36475
    },
    {
      "entropy": 5.510169267654419,
      "epoch": 3.309143686502177,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039267664790835235,
      "loss": 5.1162,
      "mean_token_accuracy": 0.1998162344098091,
      "num_tokens": 75701639.0,
      "step": 36480
    },
    {
      "entropy": 5.496618700027466,
      "epoch": 3.3095972423802613,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003926490684030134,
      "loss": 5.122,
      "mean_token_accuracy": 0.2006259083747864,
      "num_tokens": 75712823.0,
      "step": 36485
    },
    {
      "entropy": 5.494375467300415,
      "epoch": 3.3100507982583456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039262148646466185,
      "loss": 5.1551,
      "mean_token_accuracy": 0.1993646129965782,
      "num_tokens": 75723221.0,
      "step": 36490
    },
    {
      "entropy": 5.540036725997925,
      "epoch": 3.3105043541364294,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003925939020938679,
      "loss": 5.1208,
      "mean_token_accuracy": 0.2029282957315445,
      "num_tokens": 75733565.0,
      "step": 36495
    },
    {
      "entropy": 5.562301063537598,
      "epoch": 3.3109579100145137,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039256631529120206,
      "loss": 5.1134,
      "mean_token_accuracy": 0.2005388095974922,
      "num_tokens": 75743511.0,
      "step": 36500
    },
    {
      "entropy": 5.588819551467895,
      "epoch": 3.311411465892598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003925387260572349,
      "loss": 5.2144,
      "mean_token_accuracy": 0.19181180596351624,
      "num_tokens": 75754593.0,
      "step": 36505
    },
    {
      "entropy": 5.4830078125,
      "epoch": 3.311865021770682,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039251113439253693,
      "loss": 5.0843,
      "mean_token_accuracy": 0.1990000620484352,
      "num_tokens": 75764534.0,
      "step": 36510
    },
    {
      "entropy": 5.45616660118103,
      "epoch": 3.3123185776487665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003924835402976788,
      "loss": 5.0604,
      "mean_token_accuracy": 0.20045273751020432,
      "num_tokens": 75774707.0,
      "step": 36515
    },
    {
      "entropy": 5.437468814849853,
      "epoch": 3.3127721335268507,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039245594377323106,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19968865513801576,
      "num_tokens": 75784900.0,
      "step": 36520
    },
    {
      "entropy": 5.572213792800904,
      "epoch": 3.3132256894049346,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003924283448197645,
      "loss": 5.2429,
      "mean_token_accuracy": 0.19057031124830245,
      "num_tokens": 75795711.0,
      "step": 36525
    },
    {
      "entropy": 5.548658657073974,
      "epoch": 3.313679245283019,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003924007434378498,
      "loss": 5.1148,
      "mean_token_accuracy": 0.19924100935459138,
      "num_tokens": 75805951.0,
      "step": 36530
    },
    {
      "entropy": 5.519043636322022,
      "epoch": 3.314132801161103,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003923731396280579,
      "loss": 5.1186,
      "mean_token_accuracy": 0.20014672726392746,
      "num_tokens": 75816023.0,
      "step": 36535
    },
    {
      "entropy": 5.5500623226165775,
      "epoch": 3.3145863570391874,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039234553339095944,
      "loss": 5.1844,
      "mean_token_accuracy": 0.19509104192256926,
      "num_tokens": 75827395.0,
      "step": 36540
    },
    {
      "entropy": 5.572765684127807,
      "epoch": 3.315039912917271,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039231792472712554,
      "loss": 5.1035,
      "mean_token_accuracy": 0.20633053183555602,
      "num_tokens": 75836663.0,
      "step": 36545
    },
    {
      "entropy": 5.443451833724976,
      "epoch": 3.3154934687953554,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000392290313637127,
      "loss": 5.1235,
      "mean_token_accuracy": 0.19761161357164383,
      "num_tokens": 75846686.0,
      "step": 36550
    },
    {
      "entropy": 5.467913722991943,
      "epoch": 3.3159470246734397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039226270012153485,
      "loss": 5.1546,
      "mean_token_accuracy": 0.19619351923465728,
      "num_tokens": 75857386.0,
      "step": 36555
    },
    {
      "entropy": 5.5323244571685795,
      "epoch": 3.316400580551524,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003922350841809203,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19865879267454148,
      "num_tokens": 75867003.0,
      "step": 36560
    },
    {
      "entropy": 5.564442682266235,
      "epoch": 3.3168541364296082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003922074658158543,
      "loss": 5.2029,
      "mean_token_accuracy": 0.19961595982313157,
      "num_tokens": 75878155.0,
      "step": 36565
    },
    {
      "entropy": 5.502031469345093,
      "epoch": 3.3173076923076925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039217984502690797,
      "loss": 5.1244,
      "mean_token_accuracy": 0.19936110824346542,
      "num_tokens": 75888478.0,
      "step": 36570
    },
    {
      "entropy": 5.486509037017822,
      "epoch": 3.3177612481857763,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039215222181465263,
      "loss": 5.0704,
      "mean_token_accuracy": 0.2029857248067856,
      "num_tokens": 75899541.0,
      "step": 36575
    },
    {
      "entropy": 5.525616455078125,
      "epoch": 3.3182148040638606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039212459617965956,
      "loss": 5.1239,
      "mean_token_accuracy": 0.19666628986597062,
      "num_tokens": 75910175.0,
      "step": 36580
    },
    {
      "entropy": 5.462683153152466,
      "epoch": 3.318668359941945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003920969681224999,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19486008435487748,
      "num_tokens": 75919713.0,
      "step": 36585
    },
    {
      "entropy": 5.57941198348999,
      "epoch": 3.319121915820029,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039206933764374515,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1934024378657341,
      "num_tokens": 75929366.0,
      "step": 36590
    },
    {
      "entropy": 5.6705399513244625,
      "epoch": 3.3195754716981134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039204170474396656,
      "loss": 5.2447,
      "mean_token_accuracy": 0.1842617869377136,
      "num_tokens": 75940284.0,
      "step": 36595
    },
    {
      "entropy": 5.446422910690307,
      "epoch": 3.3200290275761972,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003920140694237358,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19800475835800171,
      "num_tokens": 75950752.0,
      "step": 36600
    },
    {
      "entropy": 5.514886379241943,
      "epoch": 3.3204825834542815,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003919864316836242,
      "loss": 5.1949,
      "mean_token_accuracy": 0.18708604872226714,
      "num_tokens": 75961308.0,
      "step": 36605
    },
    {
      "entropy": 5.5742271900177,
      "epoch": 3.3209361393323658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003919587915242034,
      "loss": 5.1284,
      "mean_token_accuracy": 0.2043709009885788,
      "num_tokens": 75972317.0,
      "step": 36610
    },
    {
      "entropy": 5.510120248794555,
      "epoch": 3.32138969521045,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00039193114894604504,
      "loss": 5.106,
      "mean_token_accuracy": 0.20850395709276198,
      "num_tokens": 75982769.0,
      "step": 36615
    },
    {
      "entropy": 5.467798566818237,
      "epoch": 3.3218432510885343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039190350394972074,
      "loss": 5.0916,
      "mean_token_accuracy": 0.20211703777313234,
      "num_tokens": 75992209.0,
      "step": 36620
    },
    {
      "entropy": 5.57715630531311,
      "epoch": 3.322296806966618,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003918758565358021,
      "loss": 5.1928,
      "mean_token_accuracy": 0.20293417274951936,
      "num_tokens": 76001749.0,
      "step": 36625
    },
    {
      "entropy": 5.53684778213501,
      "epoch": 3.3227503628447024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000391848206704861,
      "loss": 5.1172,
      "mean_token_accuracy": 0.1984371542930603,
      "num_tokens": 76011326.0,
      "step": 36630
    },
    {
      "entropy": 5.411845350265503,
      "epoch": 3.3232039187227866,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039182055445746917,
      "loss": 5.0447,
      "mean_token_accuracy": 0.201533442735672,
      "num_tokens": 76021527.0,
      "step": 36635
    },
    {
      "entropy": 5.488740730285644,
      "epoch": 3.323657474600871,
      "grad_norm": 1.0,
      "learning_rate": 0.00039179289979419853,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20681499391794206,
      "num_tokens": 76031217.0,
      "step": 36640
    },
    {
      "entropy": 5.5673072814941404,
      "epoch": 3.324111030478955,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039176524271562096,
      "loss": 5.2318,
      "mean_token_accuracy": 0.19180500507354736,
      "num_tokens": 76041936.0,
      "step": 36645
    },
    {
      "entropy": 5.513344478607178,
      "epoch": 3.324564586357039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039173758322230847,
      "loss": 5.185,
      "mean_token_accuracy": 0.19792769402265548,
      "num_tokens": 76052431.0,
      "step": 36650
    },
    {
      "entropy": 5.495897006988526,
      "epoch": 3.3250181422351233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003917099213148329,
      "loss": 5.035,
      "mean_token_accuracy": 0.2072020173072815,
      "num_tokens": 76061589.0,
      "step": 36655
    },
    {
      "entropy": 5.602727174758911,
      "epoch": 3.3254716981132075,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039168225699376646,
      "loss": 5.1647,
      "mean_token_accuracy": 0.20014461129903793,
      "num_tokens": 76071642.0,
      "step": 36660
    },
    {
      "entropy": 5.533149480819702,
      "epoch": 3.325925253991292,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039165459025968126,
      "loss": 5.0685,
      "mean_token_accuracy": 0.2047611504793167,
      "num_tokens": 76081084.0,
      "step": 36665
    },
    {
      "entropy": 5.501008081436157,
      "epoch": 3.326378809869376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003916269211131494,
      "loss": 5.1157,
      "mean_token_accuracy": 0.1982099324464798,
      "num_tokens": 76091125.0,
      "step": 36670
    },
    {
      "entropy": 5.605236387252807,
      "epoch": 3.3268323657474603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003915992495547431,
      "loss": 5.2129,
      "mean_token_accuracy": 0.18999056667089462,
      "num_tokens": 76101452.0,
      "step": 36675
    },
    {
      "entropy": 5.4947274684906,
      "epoch": 3.327285921625544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039157157558503456,
      "loss": 5.1077,
      "mean_token_accuracy": 0.1929059699177742,
      "num_tokens": 76112034.0,
      "step": 36680
    },
    {
      "entropy": 5.507038402557373,
      "epoch": 3.3277394775036284,
      "grad_norm": 1.125,
      "learning_rate": 0.00039154389920459615,
      "loss": 5.1219,
      "mean_token_accuracy": 0.20054849833250046,
      "num_tokens": 76122395.0,
      "step": 36685
    },
    {
      "entropy": 5.5710373878479,
      "epoch": 3.3281930333817127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003915162204140002,
      "loss": 5.1501,
      "mean_token_accuracy": 0.1950625881552696,
      "num_tokens": 76133622.0,
      "step": 36690
    },
    {
      "entropy": 5.518463563919068,
      "epoch": 3.328646589259797,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039148853921381916,
      "loss": 5.1105,
      "mean_token_accuracy": 0.19836909621953963,
      "num_tokens": 76143757.0,
      "step": 36695
    },
    {
      "entropy": 5.657582378387451,
      "epoch": 3.3291001451378808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003914608556046254,
      "loss": 5.372,
      "mean_token_accuracy": 0.18674024641513826,
      "num_tokens": 76154943.0,
      "step": 36700
    },
    {
      "entropy": 5.523778295516967,
      "epoch": 3.329553701015965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003914331695869914,
      "loss": 5.1389,
      "mean_token_accuracy": 0.19150673151016234,
      "num_tokens": 76165227.0,
      "step": 36705
    },
    {
      "entropy": 5.540172910690307,
      "epoch": 3.3300072568940493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003914054811614899,
      "loss": 5.12,
      "mean_token_accuracy": 0.19703573137521743,
      "num_tokens": 76174127.0,
      "step": 36710
    },
    {
      "entropy": 5.573985528945923,
      "epoch": 3.3304608127721336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003913777903286934,
      "loss": 5.2398,
      "mean_token_accuracy": 0.19442391246557236,
      "num_tokens": 76183518.0,
      "step": 36715
    },
    {
      "entropy": 5.509686994552612,
      "epoch": 3.330914368650218,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003913500970891745,
      "loss": 5.1832,
      "mean_token_accuracy": 0.18895311057567596,
      "num_tokens": 76193865.0,
      "step": 36720
    },
    {
      "entropy": 5.55040397644043,
      "epoch": 3.331367924528302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003913224014435058,
      "loss": 5.1855,
      "mean_token_accuracy": 0.1926089957356453,
      "num_tokens": 76204273.0,
      "step": 36725
    },
    {
      "entropy": 5.530892419815063,
      "epoch": 3.331821480406386,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039129470339226044,
      "loss": 5.0229,
      "mean_token_accuracy": 0.21113174855709077,
      "num_tokens": 76213706.0,
      "step": 36730
    },
    {
      "entropy": 5.5476133823394775,
      "epoch": 3.33227503628447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039126700293601086,
      "loss": 5.2142,
      "mean_token_accuracy": 0.1969839185476303,
      "num_tokens": 76224484.0,
      "step": 36735
    },
    {
      "entropy": 5.553557825088501,
      "epoch": 3.3327285921625545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003912393000753299,
      "loss": 5.2097,
      "mean_token_accuracy": 0.18694861829280854,
      "num_tokens": 76236004.0,
      "step": 36740
    },
    {
      "entropy": 5.552272462844849,
      "epoch": 3.3331821480406387,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039121159481079077,
      "loss": 5.1031,
      "mean_token_accuracy": 0.2001710206270218,
      "num_tokens": 76245109.0,
      "step": 36745
    },
    {
      "entropy": 5.553607416152954,
      "epoch": 3.3336357039187225,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039118388714296626,
      "loss": 5.0991,
      "mean_token_accuracy": 0.19692620784044265,
      "num_tokens": 76255154.0,
      "step": 36750
    },
    {
      "entropy": 5.498202991485596,
      "epoch": 3.334089259796807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003911561770724292,
      "loss": 5.0609,
      "mean_token_accuracy": 0.1974372923374176,
      "num_tokens": 76266014.0,
      "step": 36755
    },
    {
      "entropy": 5.389339733123779,
      "epoch": 3.334542815674891,
      "grad_norm": 1.03125,
      "learning_rate": 0.000391128464599753,
      "loss": 5.0427,
      "mean_token_accuracy": 0.2034364640712738,
      "num_tokens": 76277593.0,
      "step": 36760
    },
    {
      "entropy": 5.453934574127198,
      "epoch": 3.3349963715529753,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003911007497255104,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20605140775442124,
      "num_tokens": 76287816.0,
      "step": 36765
    },
    {
      "entropy": 5.5983133792877195,
      "epoch": 3.3354499274310596,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003910730324502748,
      "loss": 5.1615,
      "mean_token_accuracy": 0.1937492236495018,
      "num_tokens": 76298299.0,
      "step": 36770
    },
    {
      "entropy": 5.545416831970215,
      "epoch": 3.335903483309144,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003910453127746192,
      "loss": 5.1772,
      "mean_token_accuracy": 0.1867833361029625,
      "num_tokens": 76308747.0,
      "step": 36775
    },
    {
      "entropy": 5.4666211128234865,
      "epoch": 3.3363570391872277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003910175906991171,
      "loss": 5.0393,
      "mean_token_accuracy": 0.20022257417440414,
      "num_tokens": 76318923.0,
      "step": 36780
    },
    {
      "entropy": 5.521806526184082,
      "epoch": 3.336810595065312,
      "grad_norm": 1.125,
      "learning_rate": 0.0003909898662243415,
      "loss": 5.1611,
      "mean_token_accuracy": 0.19637758582830428,
      "num_tokens": 76327985.0,
      "step": 36785
    },
    {
      "entropy": 5.535371255874634,
      "epoch": 3.3372641509433962,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039096213935086604,
      "loss": 5.244,
      "mean_token_accuracy": 0.19118641763925553,
      "num_tokens": 76338933.0,
      "step": 36790
    },
    {
      "entropy": 5.694914388656616,
      "epoch": 3.3377177068214805,
      "grad_norm": 1.125,
      "learning_rate": 0.00039093441007926396,
      "loss": 5.2678,
      "mean_token_accuracy": 0.19251490831375123,
      "num_tokens": 76349601.0,
      "step": 36795
    },
    {
      "entropy": 5.556808280944824,
      "epoch": 3.3381712626995648,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039090667841010874,
      "loss": 5.11,
      "mean_token_accuracy": 0.2024338349699974,
      "num_tokens": 76359968.0,
      "step": 36800
    },
    {
      "entropy": 5.4624120712280275,
      "epoch": 3.3386248185776486,
      "grad_norm": 1.125,
      "learning_rate": 0.0003908789443439737,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20792636275291443,
      "num_tokens": 76369864.0,
      "step": 36805
    },
    {
      "entropy": 5.481876420974731,
      "epoch": 3.339078374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003908512078814329,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20820347368717193,
      "num_tokens": 76379158.0,
      "step": 36810
    },
    {
      "entropy": 5.469750022888183,
      "epoch": 3.339531930333817,
      "grad_norm": 1.125,
      "learning_rate": 0.00039082346902305937,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20335751622915268,
      "num_tokens": 76388798.0,
      "step": 36815
    },
    {
      "entropy": 5.469610929489136,
      "epoch": 3.3399854862119014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000390795727769427,
      "loss": 5.079,
      "mean_token_accuracy": 0.20378379225730897,
      "num_tokens": 76399315.0,
      "step": 36820
    },
    {
      "entropy": 5.513772964477539,
      "epoch": 3.3404390420899857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003907679841211095,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19649723619222642,
      "num_tokens": 76408841.0,
      "step": 36825
    },
    {
      "entropy": 5.5668257713317875,
      "epoch": 3.3408925979680695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003907402380786806,
      "loss": 5.2053,
      "mean_token_accuracy": 0.19104342609643937,
      "num_tokens": 76420055.0,
      "step": 36830
    },
    {
      "entropy": 5.472582578659058,
      "epoch": 3.3413461538461537,
      "grad_norm": 1.25,
      "learning_rate": 0.0003907124896427141,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20389324724674224,
      "num_tokens": 76430921.0,
      "step": 36835
    },
    {
      "entropy": 5.502431058883667,
      "epoch": 3.341799709724238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003906847388137837,
      "loss": 5.1202,
      "mean_token_accuracy": 0.19595243483781816,
      "num_tokens": 76441216.0,
      "step": 36840
    },
    {
      "entropy": 5.555234718322754,
      "epoch": 3.3422532656023223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003906569855924636,
      "loss": 5.1653,
      "mean_token_accuracy": 0.20047637224197387,
      "num_tokens": 76451940.0,
      "step": 36845
    },
    {
      "entropy": 5.553495979309082,
      "epoch": 3.3427068214804065,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003906292299793275,
      "loss": 5.1339,
      "mean_token_accuracy": 0.2059248685836792,
      "num_tokens": 76461701.0,
      "step": 36850
    },
    {
      "entropy": 5.543405437469483,
      "epoch": 3.3431603773584904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039060147197494943,
      "loss": 5.2429,
      "mean_token_accuracy": 0.1899789273738861,
      "num_tokens": 76473658.0,
      "step": 36855
    },
    {
      "entropy": 5.513333082199097,
      "epoch": 3.3436139332365746,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039057371157990347,
      "loss": 5.111,
      "mean_token_accuracy": 0.19875727742910385,
      "num_tokens": 76483821.0,
      "step": 36860
    },
    {
      "entropy": 5.529412841796875,
      "epoch": 3.344067489114659,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039054594879476374,
      "loss": 5.1453,
      "mean_token_accuracy": 0.19338738173246384,
      "num_tokens": 76494699.0,
      "step": 36865
    },
    {
      "entropy": 5.541030073165894,
      "epoch": 3.344521044992743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039051818362010433,
      "loss": 5.1166,
      "mean_token_accuracy": 0.2052502304315567,
      "num_tokens": 76504299.0,
      "step": 36870
    },
    {
      "entropy": 5.585791826248169,
      "epoch": 3.3449746008708274,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003904904160564994,
      "loss": 5.1924,
      "mean_token_accuracy": 0.19139484167099,
      "num_tokens": 76514469.0,
      "step": 36875
    },
    {
      "entropy": 5.5202208995819095,
      "epoch": 3.3454281567489117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003904626461045232,
      "loss": 5.1485,
      "mean_token_accuracy": 0.19623928368091584,
      "num_tokens": 76524666.0,
      "step": 36880
    },
    {
      "entropy": 5.462937593460083,
      "epoch": 3.3458817126269955,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039043487376475017,
      "loss": 5.0939,
      "mean_token_accuracy": 0.19625913649797438,
      "num_tokens": 76534787.0,
      "step": 36885
    },
    {
      "entropy": 5.515471506118774,
      "epoch": 3.34633526850508,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039040709903775446,
      "loss": 5.1503,
      "mean_token_accuracy": 0.19718518257141113,
      "num_tokens": 76545055.0,
      "step": 36890
    },
    {
      "entropy": 5.486792516708374,
      "epoch": 3.346788824383164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003903793219241106,
      "loss": 5.1083,
      "mean_token_accuracy": 0.20554695427417755,
      "num_tokens": 76555458.0,
      "step": 36895
    },
    {
      "entropy": 5.515979290008545,
      "epoch": 3.3472423802612483,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039035154242439294,
      "loss": 5.0847,
      "mean_token_accuracy": 0.20516700595617293,
      "num_tokens": 76566272.0,
      "step": 36900
    },
    {
      "entropy": 5.538431215286255,
      "epoch": 3.347695936139332,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003903237605391759,
      "loss": 5.1361,
      "mean_token_accuracy": 0.1960420161485672,
      "num_tokens": 76575656.0,
      "step": 36905
    },
    {
      "entropy": 5.563755178451538,
      "epoch": 3.3481494920174164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003902959762690342,
      "loss": 5.1082,
      "mean_token_accuracy": 0.1976661503314972,
      "num_tokens": 76586481.0,
      "step": 36910
    },
    {
      "entropy": 5.455104064941406,
      "epoch": 3.3486030478955007,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003902681896145423,
      "loss": 5.078,
      "mean_token_accuracy": 0.20489343404769897,
      "num_tokens": 76596023.0,
      "step": 36915
    },
    {
      "entropy": 5.461397790908814,
      "epoch": 3.349056603773585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039024040057627487,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20392300933599472,
      "num_tokens": 76605887.0,
      "step": 36920
    },
    {
      "entropy": 5.499984455108643,
      "epoch": 3.349510159651669,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003902126091548066,
      "loss": 5.0332,
      "mean_token_accuracy": 0.20611436814069747,
      "num_tokens": 76616007.0,
      "step": 36925
    },
    {
      "entropy": 5.545163869857788,
      "epoch": 3.3499637155297535,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039018481535071215,
      "loss": 5.1898,
      "mean_token_accuracy": 0.19064465165138245,
      "num_tokens": 76626253.0,
      "step": 36930
    },
    {
      "entropy": 5.501641130447387,
      "epoch": 3.3504172714078373,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003901570191645665,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20162162482738494,
      "num_tokens": 76636341.0,
      "step": 36935
    },
    {
      "entropy": 5.521026182174682,
      "epoch": 3.3508708272859216,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039012922059694416,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20325789749622344,
      "num_tokens": 76647616.0,
      "step": 36940
    },
    {
      "entropy": 5.469637441635132,
      "epoch": 3.351324383164006,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003901014196484203,
      "loss": 5.1114,
      "mean_token_accuracy": 0.20313892513513565,
      "num_tokens": 76658261.0,
      "step": 36945
    },
    {
      "entropy": 5.478233528137207,
      "epoch": 3.35177793904209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039007361631956976,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19392170757055283,
      "num_tokens": 76667985.0,
      "step": 36950
    },
    {
      "entropy": 5.505926990509034,
      "epoch": 3.3522314949201744,
      "grad_norm": 1.0,
      "learning_rate": 0.00039004581061096745,
      "loss": 5.0982,
      "mean_token_accuracy": 0.20050314217805862,
      "num_tokens": 76678538.0,
      "step": 36955
    },
    {
      "entropy": 5.5011537075042725,
      "epoch": 3.352685050798258,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003900180025231886,
      "loss": 5.0118,
      "mean_token_accuracy": 0.20616854578256608,
      "num_tokens": 76688052.0,
      "step": 36960
    },
    {
      "entropy": 5.472994232177735,
      "epoch": 3.3531386066763424,
      "grad_norm": 1.765625,
      "learning_rate": 0.00038999019205680796,
      "loss": 5.0822,
      "mean_token_accuracy": 0.20041365921497345,
      "num_tokens": 76699241.0,
      "step": 36965
    },
    {
      "entropy": 5.471687459945679,
      "epoch": 3.3535921625544267,
      "grad_norm": 1.140625,
      "learning_rate": 0.000389962379212401,
      "loss": 5.1056,
      "mean_token_accuracy": 0.1994658440351486,
      "num_tokens": 76708979.0,
      "step": 36970
    },
    {
      "entropy": 5.539930725097657,
      "epoch": 3.354045718432511,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003899345639905426,
      "loss": 5.0794,
      "mean_token_accuracy": 0.2023966372013092,
      "num_tokens": 76719330.0,
      "step": 36975
    },
    {
      "entropy": 5.525934219360352,
      "epoch": 3.3544992743105952,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003899067463918083,
      "loss": 5.2177,
      "mean_token_accuracy": 0.19196683764457703,
      "num_tokens": 76730210.0,
      "step": 36980
    },
    {
      "entropy": 5.613546419143677,
      "epoch": 3.354952830188679,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038987892641677307,
      "loss": 5.2117,
      "mean_token_accuracy": 0.1917256936430931,
      "num_tokens": 76739977.0,
      "step": 36985
    },
    {
      "entropy": 5.489967584609985,
      "epoch": 3.3554063860667633,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003898511040660124,
      "loss": 5.0805,
      "mean_token_accuracy": 0.1994861289858818,
      "num_tokens": 76750200.0,
      "step": 36990
    },
    {
      "entropy": 5.552764749526977,
      "epoch": 3.3558599419448476,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003898232793401017,
      "loss": 5.2312,
      "mean_token_accuracy": 0.18803087919950484,
      "num_tokens": 76761420.0,
      "step": 36995
    },
    {
      "entropy": 5.466790676116943,
      "epoch": 3.356313497822932,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038979545223961627,
      "loss": 5.0818,
      "mean_token_accuracy": 0.20069303065538407,
      "num_tokens": 76771872.0,
      "step": 37000
    },
    {
      "entropy": 5.48279504776001,
      "epoch": 3.356767053701016,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038976762276513163,
      "loss": 5.0585,
      "mean_token_accuracy": 0.20509777069091797,
      "num_tokens": 76782303.0,
      "step": 37005
    },
    {
      "entropy": 5.573491859436035,
      "epoch": 3.3572206095791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003897397909172232,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18956418335437775,
      "num_tokens": 76792595.0,
      "step": 37010
    },
    {
      "entropy": 5.543105792999268,
      "epoch": 3.3576741654571842,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003897119566964669,
      "loss": 5.1469,
      "mean_token_accuracy": 0.20088306218385696,
      "num_tokens": 76802577.0,
      "step": 37015
    },
    {
      "entropy": 5.51154088973999,
      "epoch": 3.3581277213352685,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003896841201034379,
      "loss": 5.081,
      "mean_token_accuracy": 0.20191854536533355,
      "num_tokens": 76812686.0,
      "step": 37020
    },
    {
      "entropy": 5.435963773727417,
      "epoch": 3.3585812772133528,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003896562811387121,
      "loss": 5.006,
      "mean_token_accuracy": 0.20913563519716263,
      "num_tokens": 76823641.0,
      "step": 37025
    },
    {
      "entropy": 5.550856351852417,
      "epoch": 3.359034833091437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038962843980286523,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19482512772083282,
      "num_tokens": 76834060.0,
      "step": 37030
    },
    {
      "entropy": 5.6184759616851805,
      "epoch": 3.3594883889695213,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038960059609647306,
      "loss": 5.238,
      "mean_token_accuracy": 0.19897391647100449,
      "num_tokens": 76845078.0,
      "step": 37035
    },
    {
      "entropy": 5.561081075668335,
      "epoch": 3.359941944847605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003895727500201113,
      "loss": 5.2226,
      "mean_token_accuracy": 0.1954079419374466,
      "num_tokens": 76856539.0,
      "step": 37040
    },
    {
      "entropy": 5.583696985244751,
      "epoch": 3.3603955007256894,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038954490157435583,
      "loss": 5.2157,
      "mean_token_accuracy": 0.19412108063697814,
      "num_tokens": 76866941.0,
      "step": 37045
    },
    {
      "entropy": 5.578017234802246,
      "epoch": 3.3608490566037736,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038951705075978264,
      "loss": 5.1042,
      "mean_token_accuracy": 0.2019186347723007,
      "num_tokens": 76878205.0,
      "step": 37050
    },
    {
      "entropy": 5.545546340942383,
      "epoch": 3.361302612481858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003894891975769677,
      "loss": 5.1565,
      "mean_token_accuracy": 0.19142176657915116,
      "num_tokens": 76888609.0,
      "step": 37055
    },
    {
      "entropy": 5.483619689941406,
      "epoch": 3.3617561683599417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003894613420264869,
      "loss": 5.1265,
      "mean_token_accuracy": 0.19646851271390914,
      "num_tokens": 76899637.0,
      "step": 37060
    },
    {
      "entropy": 5.502088975906372,
      "epoch": 3.362209724238026,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038943348410891643,
      "loss": 5.1592,
      "mean_token_accuracy": 0.19455767124891282,
      "num_tokens": 76911186.0,
      "step": 37065
    },
    {
      "entropy": 5.505025959014892,
      "epoch": 3.3626632801161103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003894056238248324,
      "loss": 5.0923,
      "mean_token_accuracy": 0.20039409548044204,
      "num_tokens": 76921704.0,
      "step": 37070
    },
    {
      "entropy": 5.434869766235352,
      "epoch": 3.3631168359941945,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003893777611748108,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20511480271816254,
      "num_tokens": 76932362.0,
      "step": 37075
    },
    {
      "entropy": 5.55811824798584,
      "epoch": 3.363570391872279,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038934989615942807,
      "loss": 5.2496,
      "mean_token_accuracy": 0.19198515117168427,
      "num_tokens": 76942581.0,
      "step": 37080
    },
    {
      "entropy": 5.538443517684937,
      "epoch": 3.364023947750363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003893220287792603,
      "loss": 5.1244,
      "mean_token_accuracy": 0.19937242716550826,
      "num_tokens": 76952912.0,
      "step": 37085
    },
    {
      "entropy": 5.542355346679687,
      "epoch": 3.364477503628447,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038929415903488383,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1963920086622238,
      "num_tokens": 76963320.0,
      "step": 37090
    },
    {
      "entropy": 5.508199644088745,
      "epoch": 3.364931059506531,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000389266286926875,
      "loss": 5.0118,
      "mean_token_accuracy": 0.2078142359852791,
      "num_tokens": 76973448.0,
      "step": 37095
    },
    {
      "entropy": 5.418643856048584,
      "epoch": 3.3653846153846154,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038923841245581026,
      "loss": 5.079,
      "mean_token_accuracy": 0.20701869428157807,
      "num_tokens": 76983671.0,
      "step": 37100
    },
    {
      "entropy": 5.535303211212158,
      "epoch": 3.3658381712626997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038921053562226607,
      "loss": 5.2257,
      "mean_token_accuracy": 0.1881590873003006,
      "num_tokens": 76994345.0,
      "step": 37105
    },
    {
      "entropy": 5.492063760757446,
      "epoch": 3.3662917271407835,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038918265642681884,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19625719934701918,
      "num_tokens": 77004809.0,
      "step": 37110
    },
    {
      "entropy": 5.479815101623535,
      "epoch": 3.3667452830188678,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038915477487004525,
      "loss": 5.1255,
      "mean_token_accuracy": 0.19753594249486922,
      "num_tokens": 77015505.0,
      "step": 37115
    },
    {
      "entropy": 5.461754608154297,
      "epoch": 3.367198838896952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038912689095252183,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21328991055488586,
      "num_tokens": 77025572.0,
      "step": 37120
    },
    {
      "entropy": 5.605990791320801,
      "epoch": 3.3676523947750363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003890990046748251,
      "loss": 5.2684,
      "mean_token_accuracy": 0.18388172686100007,
      "num_tokens": 77037273.0,
      "step": 37125
    },
    {
      "entropy": 5.558523845672608,
      "epoch": 3.3681059506531206,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038907111603753196,
      "loss": 5.192,
      "mean_token_accuracy": 0.192376109957695,
      "num_tokens": 77047724.0,
      "step": 37130
    },
    {
      "entropy": 5.489524936676025,
      "epoch": 3.368559506531205,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038904322504121916,
      "loss": 5.1012,
      "mean_token_accuracy": 0.1979268252849579,
      "num_tokens": 77057839.0,
      "step": 37135
    },
    {
      "entropy": 5.472234296798706,
      "epoch": 3.3690130624092887,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003890153316864632,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19152533113956452,
      "num_tokens": 77068505.0,
      "step": 37140
    },
    {
      "entropy": 5.562122631072998,
      "epoch": 3.369466618287373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038898743597384126,
      "loss": 5.1834,
      "mean_token_accuracy": 0.19614955335855483,
      "num_tokens": 77078126.0,
      "step": 37145
    },
    {
      "entropy": 5.5004466533660885,
      "epoch": 3.369920174165457,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038895953790393014,
      "loss": 5.1222,
      "mean_token_accuracy": 0.2051307588815689,
      "num_tokens": 77087802.0,
      "step": 37150
    },
    {
      "entropy": 5.572225189208984,
      "epoch": 3.3703737300435415,
      "grad_norm": 1.0,
      "learning_rate": 0.00038893163747730664,
      "loss": 5.1202,
      "mean_token_accuracy": 0.20461261570453643,
      "num_tokens": 77097901.0,
      "step": 37155
    },
    {
      "entropy": 5.512279272079468,
      "epoch": 3.3708272859216257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038890373469454794,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19884648472070693,
      "num_tokens": 77107604.0,
      "step": 37160
    },
    {
      "entropy": 5.492062568664551,
      "epoch": 3.3712808417997095,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038887582955623084,
      "loss": 5.0618,
      "mean_token_accuracy": 0.20551876872777938,
      "num_tokens": 77117410.0,
      "step": 37165
    },
    {
      "entropy": 5.571507167816162,
      "epoch": 3.371734397677794,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003888479220629326,
      "loss": 5.1447,
      "mean_token_accuracy": 0.19482154697179793,
      "num_tokens": 77129546.0,
      "step": 37170
    },
    {
      "entropy": 5.491101360321045,
      "epoch": 3.372187953555878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003888200122152304,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20542621314525605,
      "num_tokens": 77139217.0,
      "step": 37175
    },
    {
      "entropy": 5.5414121627807615,
      "epoch": 3.3726415094339623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003887921000137013,
      "loss": 5.2432,
      "mean_token_accuracy": 0.1922924265265465,
      "num_tokens": 77149694.0,
      "step": 37180
    },
    {
      "entropy": 5.497832345962524,
      "epoch": 3.3730950653120466,
      "grad_norm": 1.125,
      "learning_rate": 0.0003887641854589226,
      "loss": 5.0914,
      "mean_token_accuracy": 0.20444636344909667,
      "num_tokens": 77159291.0,
      "step": 37185
    },
    {
      "entropy": 5.4901398658752445,
      "epoch": 3.3735486211901304,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003887362685514716,
      "loss": 5.0987,
      "mean_token_accuracy": 0.19551612287759781,
      "num_tokens": 77170507.0,
      "step": 37190
    },
    {
      "entropy": 5.506915044784546,
      "epoch": 3.3740021770682147,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003887083492919254,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19945946782827378,
      "num_tokens": 77179673.0,
      "step": 37195
    },
    {
      "entropy": 5.514212799072266,
      "epoch": 3.374455732946299,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003886804276808617,
      "loss": 5.1187,
      "mean_token_accuracy": 0.20668163001537324,
      "num_tokens": 77190163.0,
      "step": 37200
    },
    {
      "entropy": 5.553794527053833,
      "epoch": 3.3749092888243832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003886525037188578,
      "loss": 5.1439,
      "mean_token_accuracy": 0.20087750256061554,
      "num_tokens": 77201498.0,
      "step": 37205
    },
    {
      "entropy": 5.518740606307984,
      "epoch": 3.3753628447024675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003886245774064911,
      "loss": 5.065,
      "mean_token_accuracy": 0.2028060585260391,
      "num_tokens": 77211475.0,
      "step": 37210
    },
    {
      "entropy": 5.504223394393921,
      "epoch": 3.3758164005805513,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038859664874433923,
      "loss": 5.1065,
      "mean_token_accuracy": 0.2006707787513733,
      "num_tokens": 77221972.0,
      "step": 37215
    },
    {
      "entropy": 5.532559537887574,
      "epoch": 3.3762699564586356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038856871773297965,
      "loss": 5.1374,
      "mean_token_accuracy": 0.19480928033590317,
      "num_tokens": 77232701.0,
      "step": 37220
    },
    {
      "entropy": 5.52751202583313,
      "epoch": 3.37672351233672,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038854078437299005,
      "loss": 5.0823,
      "mean_token_accuracy": 0.2076243206858635,
      "num_tokens": 77243242.0,
      "step": 37225
    },
    {
      "entropy": 5.471125316619873,
      "epoch": 3.377177068214804,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003885128486649482,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20395849645137787,
      "num_tokens": 77254279.0,
      "step": 37230
    },
    {
      "entropy": 5.53330135345459,
      "epoch": 3.3776306240928884,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038848491060943164,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19768979549407958,
      "num_tokens": 77265483.0,
      "step": 37235
    },
    {
      "entropy": 5.458680438995361,
      "epoch": 3.3780841799709727,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003884569702070182,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20143570899963378,
      "num_tokens": 77275848.0,
      "step": 37240
    },
    {
      "entropy": 5.417066049575806,
      "epoch": 3.3785377358490565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038842902745828573,
      "loss": 5.0579,
      "mean_token_accuracy": 0.1997240498661995,
      "num_tokens": 77285943.0,
      "step": 37245
    },
    {
      "entropy": 5.466315078735351,
      "epoch": 3.3789912917271407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003884010823638121,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20199064910411835,
      "num_tokens": 77295276.0,
      "step": 37250
    },
    {
      "entropy": 5.5353090286254885,
      "epoch": 3.379444847605225,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003883731349241751,
      "loss": 5.1081,
      "mean_token_accuracy": 0.2012190654873848,
      "num_tokens": 77305713.0,
      "step": 37255
    },
    {
      "entropy": 5.537188291549683,
      "epoch": 3.3798984034833093,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003883451851399528,
      "loss": 5.2346,
      "mean_token_accuracy": 0.19956114441156386,
      "num_tokens": 77316717.0,
      "step": 37260
    },
    {
      "entropy": 5.501924467086792,
      "epoch": 3.380351959361393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038831723301172325,
      "loss": 5.044,
      "mean_token_accuracy": 0.2036737397313118,
      "num_tokens": 77327763.0,
      "step": 37265
    },
    {
      "entropy": 5.5624302387237545,
      "epoch": 3.3808055152394774,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003882892785400644,
      "loss": 5.1746,
      "mean_token_accuracy": 0.19611739367246628,
      "num_tokens": 77337495.0,
      "step": 37270
    },
    {
      "entropy": 5.551480960845947,
      "epoch": 3.3812590711175616,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038826132172555444,
      "loss": 5.1928,
      "mean_token_accuracy": 0.19415640234947204,
      "num_tokens": 77348849.0,
      "step": 37275
    },
    {
      "entropy": 5.530072355270386,
      "epoch": 3.381712626995646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003882333625687714,
      "loss": 5.1544,
      "mean_token_accuracy": 0.1939098969101906,
      "num_tokens": 77359634.0,
      "step": 37280
    },
    {
      "entropy": 5.526908683776855,
      "epoch": 3.38216618287373,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038820540107029365,
      "loss": 5.0721,
      "mean_token_accuracy": 0.20140738785266876,
      "num_tokens": 77369342.0,
      "step": 37285
    },
    {
      "entropy": 5.553524160385132,
      "epoch": 3.3826197387518144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003881774372306993,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1939724862575531,
      "num_tokens": 77379963.0,
      "step": 37290
    },
    {
      "entropy": 5.470476055145264,
      "epoch": 3.3830732946298983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003881494710505667,
      "loss": 5.0606,
      "mean_token_accuracy": 0.19895974695682525,
      "num_tokens": 77390135.0,
      "step": 37295
    },
    {
      "entropy": 5.596695756912231,
      "epoch": 3.3835268505079825,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038812150253047427,
      "loss": 5.214,
      "mean_token_accuracy": 0.19677176475524902,
      "num_tokens": 77400151.0,
      "step": 37300
    },
    {
      "entropy": 5.488976240158081,
      "epoch": 3.383980406386067,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003880935316710003,
      "loss": 5.1106,
      "mean_token_accuracy": 0.20614362955093385,
      "num_tokens": 77410723.0,
      "step": 37305
    },
    {
      "entropy": 5.490431022644043,
      "epoch": 3.384433962264151,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003880655584727233,
      "loss": 5.0415,
      "mean_token_accuracy": 0.21044483482837678,
      "num_tokens": 77421213.0,
      "step": 37310
    },
    {
      "entropy": 5.553554344177246,
      "epoch": 3.384887518142235,
      "grad_norm": 1.125,
      "learning_rate": 0.0003880375829362217,
      "loss": 5.154,
      "mean_token_accuracy": 0.19719438552856444,
      "num_tokens": 77431705.0,
      "step": 37315
    },
    {
      "entropy": 5.512286376953125,
      "epoch": 3.385341074020319,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003880096050620741,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1995377466082573,
      "num_tokens": 77442120.0,
      "step": 37320
    },
    {
      "entropy": 5.518416929244995,
      "epoch": 3.3857946298984034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003879816248508591,
      "loss": 5.078,
      "mean_token_accuracy": 0.19902192652225495,
      "num_tokens": 77451992.0,
      "step": 37325
    },
    {
      "entropy": 5.5611495018005375,
      "epoch": 3.3862481857764877,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038795364230315523,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1835303410887718,
      "num_tokens": 77462830.0,
      "step": 37330
    },
    {
      "entropy": 5.575249767303466,
      "epoch": 3.386701741654572,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003879256574195414,
      "loss": 5.1467,
      "mean_token_accuracy": 0.20170280784368516,
      "num_tokens": 77472617.0,
      "step": 37335
    },
    {
      "entropy": 5.493180656433106,
      "epoch": 3.387155297532656,
      "grad_norm": 1.25,
      "learning_rate": 0.00038789767020059614,
      "loss": 5.0531,
      "mean_token_accuracy": 0.20360539704561234,
      "num_tokens": 77483430.0,
      "step": 37340
    },
    {
      "entropy": 5.43919677734375,
      "epoch": 3.38760885341074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038786968064689816,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20362301766872407,
      "num_tokens": 77493038.0,
      "step": 37345
    },
    {
      "entropy": 5.531312561035156,
      "epoch": 3.3880624092888243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003878416887590265,
      "loss": 5.1565,
      "mean_token_accuracy": 0.1892839640378952,
      "num_tokens": 77502816.0,
      "step": 37350
    },
    {
      "entropy": 5.464830636978149,
      "epoch": 3.3885159651669086,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038781369453756004,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20591408461332322,
      "num_tokens": 77513026.0,
      "step": 37355
    },
    {
      "entropy": 5.5470422267913815,
      "epoch": 3.388969521044993,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003877856979830775,
      "loss": 5.119,
      "mean_token_accuracy": 0.19673834294080733,
      "num_tokens": 77523217.0,
      "step": 37360
    },
    {
      "entropy": 5.5295497417449955,
      "epoch": 3.389423076923077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038775769909615805,
      "loss": 5.1917,
      "mean_token_accuracy": 0.1962285041809082,
      "num_tokens": 77533339.0,
      "step": 37365
    },
    {
      "entropy": 5.4927207946777346,
      "epoch": 3.389876632801161,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038772969787738074,
      "loss": 5.09,
      "mean_token_accuracy": 0.20470571368932725,
      "num_tokens": 77543438.0,
      "step": 37370
    },
    {
      "entropy": 5.537673664093018,
      "epoch": 3.390330188679245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003877016943273244,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19735153168439865,
      "num_tokens": 77552825.0,
      "step": 37375
    },
    {
      "entropy": 5.499099588394165,
      "epoch": 3.3907837445573294,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038767368844656836,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19433834850788118,
      "num_tokens": 77563107.0,
      "step": 37380
    },
    {
      "entropy": 5.560703086853027,
      "epoch": 3.3912373004354137,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038764568023569177,
      "loss": 5.1657,
      "mean_token_accuracy": 0.18966831713914872,
      "num_tokens": 77573312.0,
      "step": 37385
    },
    {
      "entropy": 5.497561311721801,
      "epoch": 3.391690856313498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003876176696952737,
      "loss": 5.1287,
      "mean_token_accuracy": 0.1949734255671501,
      "num_tokens": 77583742.0,
      "step": 37390
    },
    {
      "entropy": 5.603786325454712,
      "epoch": 3.3921444121915822,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038758965682589357,
      "loss": 5.1407,
      "mean_token_accuracy": 0.19158224612474442,
      "num_tokens": 77593667.0,
      "step": 37395
    },
    {
      "entropy": 5.471639108657837,
      "epoch": 3.392597968069666,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003875616416281307,
      "loss": 5.0439,
      "mean_token_accuracy": 0.20431844145059586,
      "num_tokens": 77604249.0,
      "step": 37400
    },
    {
      "entropy": 5.489247894287109,
      "epoch": 3.3930515239477503,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038753362410256435,
      "loss": 5.0836,
      "mean_token_accuracy": 0.20276067703962325,
      "num_tokens": 77615095.0,
      "step": 37405
    },
    {
      "entropy": 5.463066911697387,
      "epoch": 3.3935050798258346,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003875056042497739,
      "loss": 5.0631,
      "mean_token_accuracy": 0.20607273131608964,
      "num_tokens": 77625182.0,
      "step": 37410
    },
    {
      "entropy": 5.491796779632568,
      "epoch": 3.393958635703919,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038747758207033903,
      "loss": 5.0878,
      "mean_token_accuracy": 0.20194550156593322,
      "num_tokens": 77635423.0,
      "step": 37415
    },
    {
      "entropy": 5.417299413681031,
      "epoch": 3.3944121915820027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003874495575648391,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20826620757579803,
      "num_tokens": 77645262.0,
      "step": 37420
    },
    {
      "entropy": 5.507989835739136,
      "epoch": 3.394865747460087,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038742153073385356,
      "loss": 5.1726,
      "mean_token_accuracy": 0.19135786592960358,
      "num_tokens": 77657647.0,
      "step": 37425
    },
    {
      "entropy": 5.574865388870239,
      "epoch": 3.3953193033381712,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038739350157796214,
      "loss": 5.1222,
      "mean_token_accuracy": 0.19949911683797836,
      "num_tokens": 77668074.0,
      "step": 37430
    },
    {
      "entropy": 5.538959980010986,
      "epoch": 3.3957728592162555,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038736547009774445,
      "loss": 5.0685,
      "mean_token_accuracy": 0.19851029962301253,
      "num_tokens": 77678661.0,
      "step": 37435
    },
    {
      "entropy": 5.483286380767822,
      "epoch": 3.3962264150943398,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003873374362937803,
      "loss": 5.2027,
      "mean_token_accuracy": 0.191997691988945,
      "num_tokens": 77689206.0,
      "step": 37440
    },
    {
      "entropy": 5.507355117797852,
      "epoch": 3.396679970972424,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003873094001666493,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1964624583721161,
      "num_tokens": 77699254.0,
      "step": 37445
    },
    {
      "entropy": 5.5607099533081055,
      "epoch": 3.397133526850508,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003872813617169312,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19988471418619155,
      "num_tokens": 77709595.0,
      "step": 37450
    },
    {
      "entropy": 5.482919406890869,
      "epoch": 3.397587082728592,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038725332094520607,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20249563753604888,
      "num_tokens": 77719446.0,
      "step": 37455
    },
    {
      "entropy": 5.436529159545898,
      "epoch": 3.3980406386066764,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003872252778520536,
      "loss": 5.0955,
      "mean_token_accuracy": 0.2050565794110298,
      "num_tokens": 77729723.0,
      "step": 37460
    },
    {
      "entropy": 5.478138637542725,
      "epoch": 3.3984941944847606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003871972324380537,
      "loss": 5.0644,
      "mean_token_accuracy": 0.19761668145656586,
      "num_tokens": 77739380.0,
      "step": 37465
    },
    {
      "entropy": 5.554465913772583,
      "epoch": 3.3989477503628445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003871691847037865,
      "loss": 5.137,
      "mean_token_accuracy": 0.2049694061279297,
      "num_tokens": 77750683.0,
      "step": 37470
    },
    {
      "entropy": 5.529434633255005,
      "epoch": 3.3994013062409287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038714113464983205,
      "loss": 5.1618,
      "mean_token_accuracy": 0.19187997728586198,
      "num_tokens": 77761951.0,
      "step": 37475
    },
    {
      "entropy": 5.521352052688599,
      "epoch": 3.399854862119013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003871130822767702,
      "loss": 5.1532,
      "mean_token_accuracy": 0.1975140690803528,
      "num_tokens": 77772165.0,
      "step": 37480
    },
    {
      "entropy": 5.575578784942627,
      "epoch": 3.4003084179970973,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003870850275851814,
      "loss": 5.1952,
      "mean_token_accuracy": 0.1925129398703575,
      "num_tokens": 77781916.0,
      "step": 37485
    },
    {
      "entropy": 5.490322113037109,
      "epoch": 3.4007619738751815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038705697057564556,
      "loss": 5.0668,
      "mean_token_accuracy": 0.20041995197534562,
      "num_tokens": 77791397.0,
      "step": 37490
    },
    {
      "entropy": 5.431194067001343,
      "epoch": 3.401215529753266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000387028911248743,
      "loss": 5.0775,
      "mean_token_accuracy": 0.20251262485980986,
      "num_tokens": 77801694.0,
      "step": 37495
    },
    {
      "entropy": 5.532680416107178,
      "epoch": 3.4016690856313496,
      "grad_norm": 1.125,
      "learning_rate": 0.00038700084960505414,
      "loss": 5.2034,
      "mean_token_accuracy": 0.19136641621589662,
      "num_tokens": 77813199.0,
      "step": 37500
    },
    {
      "entropy": 5.6164881706237795,
      "epoch": 3.402122641509434,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003869727856451591,
      "loss": 5.232,
      "mean_token_accuracy": 0.1981295332312584,
      "num_tokens": 77823689.0,
      "step": 37505
    },
    {
      "entropy": 5.544637680053711,
      "epoch": 3.402576197387518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038694471936963825,
      "loss": 5.0665,
      "mean_token_accuracy": 0.20692750960588455,
      "num_tokens": 77834331.0,
      "step": 37510
    },
    {
      "entropy": 5.466018629074097,
      "epoch": 3.4030297532656024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038691665077907213,
      "loss": 5.0146,
      "mean_token_accuracy": 0.19978409111499787,
      "num_tokens": 77843988.0,
      "step": 37515
    },
    {
      "entropy": 5.447806549072266,
      "epoch": 3.4034833091436867,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003868885798740412,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20267145782709123,
      "num_tokens": 77854553.0,
      "step": 37520
    },
    {
      "entropy": 5.476548576354981,
      "epoch": 3.4039368650217705,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038686050665512583,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20757026225328445,
      "num_tokens": 77864629.0,
      "step": 37525
    },
    {
      "entropy": 5.540769052505493,
      "epoch": 3.4043904208998548,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038683243112290673,
      "loss": 5.1636,
      "mean_token_accuracy": 0.19346576929092407,
      "num_tokens": 77875465.0,
      "step": 37530
    },
    {
      "entropy": 5.561075687408447,
      "epoch": 3.404843976777939,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003868043532779645,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19581180363893508,
      "num_tokens": 77885337.0,
      "step": 37535
    },
    {
      "entropy": 5.483947229385376,
      "epoch": 3.4052975326560233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038677627312087967,
      "loss": 5.0917,
      "mean_token_accuracy": 0.19887364208698272,
      "num_tokens": 77896465.0,
      "step": 37540
    },
    {
      "entropy": 5.45949068069458,
      "epoch": 3.4057510885341076,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003867481906522331,
      "loss": 5.1009,
      "mean_token_accuracy": 0.2004752054810524,
      "num_tokens": 77906214.0,
      "step": 37545
    },
    {
      "entropy": 5.56634030342102,
      "epoch": 3.4062046444121914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003867201058726054,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18715809136629105,
      "num_tokens": 77917466.0,
      "step": 37550
    },
    {
      "entropy": 5.660714149475098,
      "epoch": 3.4066582002902757,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003866920187825775,
      "loss": 5.2027,
      "mean_token_accuracy": 0.19328700602054597,
      "num_tokens": 77927648.0,
      "step": 37555
    },
    {
      "entropy": 5.470282363891601,
      "epoch": 3.40711175616836,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038666392938273015,
      "loss": 5.0362,
      "mean_token_accuracy": 0.20439703166484832,
      "num_tokens": 77938028.0,
      "step": 37560
    },
    {
      "entropy": 5.461587429046631,
      "epoch": 3.407565312046444,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003866358376736443,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20112763345241547,
      "num_tokens": 77947276.0,
      "step": 37565
    },
    {
      "entropy": 5.526230716705323,
      "epoch": 3.4080188679245285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003866077436559009,
      "loss": 5.1402,
      "mean_token_accuracy": 0.1933736801147461,
      "num_tokens": 77956791.0,
      "step": 37570
    },
    {
      "entropy": 5.564766311645508,
      "epoch": 3.4084724238026123,
      "grad_norm": 1.125,
      "learning_rate": 0.0003865796473300809,
      "loss": 5.1351,
      "mean_token_accuracy": 0.1972146824002266,
      "num_tokens": 77967402.0,
      "step": 37575
    },
    {
      "entropy": 5.624421501159668,
      "epoch": 3.4089259796806965,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003865515486967653,
      "loss": 5.2559,
      "mean_token_accuracy": 0.1919920936226845,
      "num_tokens": 77978000.0,
      "step": 37580
    },
    {
      "entropy": 5.473760795593262,
      "epoch": 3.409379535558781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038652344775653535,
      "loss": 5.038,
      "mean_token_accuracy": 0.20956139862537385,
      "num_tokens": 77987959.0,
      "step": 37585
    },
    {
      "entropy": 5.48282823562622,
      "epoch": 3.409833091436865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003864953445099721,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20337637811899184,
      "num_tokens": 77997990.0,
      "step": 37590
    },
    {
      "entropy": 5.5440438747406,
      "epoch": 3.4102866473149493,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038646723895765654,
      "loss": 5.1492,
      "mean_token_accuracy": 0.1946120724081993,
      "num_tokens": 78008831.0,
      "step": 37595
    },
    {
      "entropy": 5.576324701309204,
      "epoch": 3.4107402031930336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003864391311001702,
      "loss": 5.2177,
      "mean_token_accuracy": 0.19169274270534514,
      "num_tokens": 78019338.0,
      "step": 37600
    },
    {
      "entropy": 5.571734380722046,
      "epoch": 3.4111937590711174,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003864110209380942,
      "loss": 5.1999,
      "mean_token_accuracy": 0.1894042059779167,
      "num_tokens": 78030700.0,
      "step": 37605
    },
    {
      "entropy": 5.517847061157227,
      "epoch": 3.4116473149492017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038638290847200997,
      "loss": 5.069,
      "mean_token_accuracy": 0.2059629514813423,
      "num_tokens": 78040190.0,
      "step": 37610
    },
    {
      "entropy": 5.488764476776123,
      "epoch": 3.412100870827286,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003863547937024987,
      "loss": 5.0851,
      "mean_token_accuracy": 0.20672218203544618,
      "num_tokens": 78050847.0,
      "step": 37615
    },
    {
      "entropy": 5.578703355789185,
      "epoch": 3.4125544267053702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000386326676630142,
      "loss": 5.1355,
      "mean_token_accuracy": 0.19954791516065598,
      "num_tokens": 78060584.0,
      "step": 37620
    },
    {
      "entropy": 5.489506721496582,
      "epoch": 3.413007982583454,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003862985572555212,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19899484217166902,
      "num_tokens": 78071189.0,
      "step": 37625
    },
    {
      "entropy": 5.498886585235596,
      "epoch": 3.4134615384615383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038627043557921785,
      "loss": 5.1013,
      "mean_token_accuracy": 0.20115175247192382,
      "num_tokens": 78082054.0,
      "step": 37630
    },
    {
      "entropy": 5.454078435897827,
      "epoch": 3.4139150943396226,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003862423116018136,
      "loss": 5.1037,
      "mean_token_accuracy": 0.19345097690820695,
      "num_tokens": 78094032.0,
      "step": 37635
    },
    {
      "entropy": 5.509957647323608,
      "epoch": 3.414368650217707,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038621418532389004,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1987249031662941,
      "num_tokens": 78104724.0,
      "step": 37640
    },
    {
      "entropy": 5.573404359817505,
      "epoch": 3.414822206095791,
      "grad_norm": 1.78125,
      "learning_rate": 0.00038618605674602867,
      "loss": 5.1958,
      "mean_token_accuracy": 0.18940706700086593,
      "num_tokens": 78115927.0,
      "step": 37645
    },
    {
      "entropy": 5.550610542297363,
      "epoch": 3.4152757619738754,
      "grad_norm": 1.125,
      "learning_rate": 0.00038615792586881145,
      "loss": 5.1608,
      "mean_token_accuracy": 0.2032097563147545,
      "num_tokens": 78126744.0,
      "step": 37650
    },
    {
      "entropy": 5.4895679473876955,
      "epoch": 3.415729317851959,
      "grad_norm": 1.125,
      "learning_rate": 0.0003861297926928199,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19941272139549254,
      "num_tokens": 78137625.0,
      "step": 37655
    },
    {
      "entropy": 5.438246870040894,
      "epoch": 3.4161828737300435,
      "grad_norm": 1.25,
      "learning_rate": 0.0003861016572186359,
      "loss": 5.045,
      "mean_token_accuracy": 0.2071939840912819,
      "num_tokens": 78147709.0,
      "step": 37660
    },
    {
      "entropy": 5.59706220626831,
      "epoch": 3.4166364296081277,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038607351944684135,
      "loss": 5.2676,
      "mean_token_accuracy": 0.19484493732452393,
      "num_tokens": 78158634.0,
      "step": 37665
    },
    {
      "entropy": 5.614883089065552,
      "epoch": 3.417089985486212,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003860453793780181,
      "loss": 5.2108,
      "mean_token_accuracy": 0.19186438024044036,
      "num_tokens": 78170279.0,
      "step": 37670
    },
    {
      "entropy": 5.504054594039917,
      "epoch": 3.417543541364296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038601723701274813,
      "loss": 5.1197,
      "mean_token_accuracy": 0.20195963680744172,
      "num_tokens": 78180209.0,
      "step": 37675
    },
    {
      "entropy": 5.513294887542725,
      "epoch": 3.41799709724238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003859890923516134,
      "loss": 5.1623,
      "mean_token_accuracy": 0.20149533450603485,
      "num_tokens": 78189937.0,
      "step": 37680
    },
    {
      "entropy": 5.48692512512207,
      "epoch": 3.4184506531204644,
      "grad_norm": 1.15625,
      "learning_rate": 0.000385960945395196,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19667260050773622,
      "num_tokens": 78199224.0,
      "step": 37685
    },
    {
      "entropy": 5.5215472221374515,
      "epoch": 3.4189042089985486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003859327961440779,
      "loss": 5.0847,
      "mean_token_accuracy": 0.20209712088108062,
      "num_tokens": 78209912.0,
      "step": 37690
    },
    {
      "entropy": 5.5375207424163815,
      "epoch": 3.419357764876633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038590464459884133,
      "loss": 5.1386,
      "mean_token_accuracy": 0.20011374801397325,
      "num_tokens": 78220388.0,
      "step": 37695
    },
    {
      "entropy": 5.557272243499756,
      "epoch": 3.419811320754717,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038587649076006844,
      "loss": 5.2341,
      "mean_token_accuracy": 0.19117405116558076,
      "num_tokens": 78231094.0,
      "step": 37700
    },
    {
      "entropy": 5.5559648990631105,
      "epoch": 3.420264876632801,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003858483346283415,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20051166266202927,
      "num_tokens": 78241111.0,
      "step": 37705
    },
    {
      "entropy": 5.555383253097534,
      "epoch": 3.4207184325108853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003858201762042428,
      "loss": 5.1147,
      "mean_token_accuracy": 0.19738731831312178,
      "num_tokens": 78251036.0,
      "step": 37710
    },
    {
      "entropy": 5.472865295410156,
      "epoch": 3.4211719883889695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003857920154883545,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1966871738433838,
      "num_tokens": 78261840.0,
      "step": 37715
    },
    {
      "entropy": 5.495642423629761,
      "epoch": 3.421625544267054,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003857638524812591,
      "loss": 5.0866,
      "mean_token_accuracy": 0.19519510865211487,
      "num_tokens": 78272539.0,
      "step": 37720
    },
    {
      "entropy": 5.5678081035614015,
      "epoch": 3.422079100145138,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038573568718353914,
      "loss": 5.0964,
      "mean_token_accuracy": 0.20061236172914504,
      "num_tokens": 78283849.0,
      "step": 37725
    },
    {
      "entropy": 5.547879981994629,
      "epoch": 3.422532656023222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003857075195957768,
      "loss": 5.1278,
      "mean_token_accuracy": 0.19398583620786666,
      "num_tokens": 78293640.0,
      "step": 37730
    },
    {
      "entropy": 5.4935546875,
      "epoch": 3.422986211901306,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003856793497185548,
      "loss": 5.0995,
      "mean_token_accuracy": 0.20217680931091309,
      "num_tokens": 78303980.0,
      "step": 37735
    },
    {
      "entropy": 5.476669788360596,
      "epoch": 3.4234397677793904,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003856511775524556,
      "loss": 5.0625,
      "mean_token_accuracy": 0.19842862635850905,
      "num_tokens": 78314233.0,
      "step": 37740
    },
    {
      "entropy": 5.550369453430176,
      "epoch": 3.4238933236574747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038562300309806193,
      "loss": 5.1015,
      "mean_token_accuracy": 0.20524923652410507,
      "num_tokens": 78324449.0,
      "step": 37745
    },
    {
      "entropy": 5.581402969360352,
      "epoch": 3.424346879535559,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003855948263559563,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19685669392347335,
      "num_tokens": 78334357.0,
      "step": 37750
    },
    {
      "entropy": 5.5562560081481935,
      "epoch": 3.424800435413643,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038556664732672147,
      "loss": 5.0951,
      "mean_token_accuracy": 0.2009773313999176,
      "num_tokens": 78344221.0,
      "step": 37755
    },
    {
      "entropy": 5.491454553604126,
      "epoch": 3.425253991291727,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038553846601094033,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20468805283308028,
      "num_tokens": 78354097.0,
      "step": 37760
    },
    {
      "entropy": 5.562901210784912,
      "epoch": 3.4257075471698113,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003855102824091954,
      "loss": 5.1905,
      "mean_token_accuracy": 0.19592179358005524,
      "num_tokens": 78363797.0,
      "step": 37765
    },
    {
      "entropy": 5.532795476913452,
      "epoch": 3.4261611030478956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003854820965220697,
      "loss": 5.0956,
      "mean_token_accuracy": 0.1967570811510086,
      "num_tokens": 78374084.0,
      "step": 37770
    },
    {
      "entropy": 5.566942739486694,
      "epoch": 3.42661465892598,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038545390835014614,
      "loss": 5.1162,
      "mean_token_accuracy": 0.1913689687848091,
      "num_tokens": 78385346.0,
      "step": 37775
    },
    {
      "entropy": 5.490192794799805,
      "epoch": 3.4270682148040637,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038542571789400754,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20009929537773133,
      "num_tokens": 78395498.0,
      "step": 37780
    },
    {
      "entropy": 5.423812580108643,
      "epoch": 3.427521770682148,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038539752515423704,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19554221779108047,
      "num_tokens": 78405792.0,
      "step": 37785
    },
    {
      "entropy": 5.563810634613037,
      "epoch": 3.427975326560232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003853693301314175,
      "loss": 5.1611,
      "mean_token_accuracy": 0.19910932779312135,
      "num_tokens": 78415672.0,
      "step": 37790
    },
    {
      "entropy": 5.488062953948974,
      "epoch": 3.4284288824383164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003853411328261321,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20360953956842423,
      "num_tokens": 78426937.0,
      "step": 37795
    },
    {
      "entropy": 5.477718830108643,
      "epoch": 3.4288824383164007,
      "grad_norm": 1.09375,
      "learning_rate": 0.000385312933238964,
      "loss": 5.125,
      "mean_token_accuracy": 0.20005601793527603,
      "num_tokens": 78436817.0,
      "step": 37800
    },
    {
      "entropy": 5.568025016784668,
      "epoch": 3.429335994194485,
      "grad_norm": 1.125,
      "learning_rate": 0.0003852847313704963,
      "loss": 5.2045,
      "mean_token_accuracy": 0.19840413182973862,
      "num_tokens": 78447348.0,
      "step": 37805
    },
    {
      "entropy": 5.551898241043091,
      "epoch": 3.429789550072569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003852565272213123,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19992495328187943,
      "num_tokens": 78457631.0,
      "step": 37810
    },
    {
      "entropy": 5.446177291870117,
      "epoch": 3.430243105950653,
      "grad_norm": 1.0,
      "learning_rate": 0.0003852283207919951,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20880455672740936,
      "num_tokens": 78468542.0,
      "step": 37815
    },
    {
      "entropy": 5.541527032852173,
      "epoch": 3.4306966618287373,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003852001120831282,
      "loss": 5.1488,
      "mean_token_accuracy": 0.1996956318616867,
      "num_tokens": 78478580.0,
      "step": 37820
    },
    {
      "entropy": 5.5202953815460205,
      "epoch": 3.4311502177068216,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00038517190109529497,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19829483926296235,
      "num_tokens": 78488477.0,
      "step": 37825
    },
    {
      "entropy": 5.559776163101196,
      "epoch": 3.4316037735849054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038514368782907864,
      "loss": 5.1348,
      "mean_token_accuracy": 0.20281215459108354,
      "num_tokens": 78498887.0,
      "step": 37830
    },
    {
      "entropy": 5.539032316207885,
      "epoch": 3.4320573294629897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038511547228506284,
      "loss": 5.0999,
      "mean_token_accuracy": 0.19404542893171312,
      "num_tokens": 78509030.0,
      "step": 37835
    },
    {
      "entropy": 5.415467214584351,
      "epoch": 3.432510885341074,
      "grad_norm": 1.125,
      "learning_rate": 0.00038508725446383097,
      "loss": 4.9604,
      "mean_token_accuracy": 0.20796728432178496,
      "num_tokens": 78518254.0,
      "step": 37840
    },
    {
      "entropy": 5.494000005722046,
      "epoch": 3.4329644412191582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003850590343659667,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20432258099317552,
      "num_tokens": 78527966.0,
      "step": 37845
    },
    {
      "entropy": 5.505051898956299,
      "epoch": 3.4334179970972425,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038503081199205353,
      "loss": 5.1913,
      "mean_token_accuracy": 0.1985347405076027,
      "num_tokens": 78537932.0,
      "step": 37850
    },
    {
      "entropy": 5.5868912696838375,
      "epoch": 3.4338715529753268,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038500258734267505,
      "loss": 5.2169,
      "mean_token_accuracy": 0.19584798961877822,
      "num_tokens": 78549213.0,
      "step": 37855
    },
    {
      "entropy": 5.482662677764893,
      "epoch": 3.4343251088534106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038497436041841517,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19875307977199555,
      "num_tokens": 78558990.0,
      "step": 37860
    },
    {
      "entropy": 5.473666191101074,
      "epoch": 3.434778664731495,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038494613121985735,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19803021997213363,
      "num_tokens": 78568496.0,
      "step": 37865
    },
    {
      "entropy": 5.482982635498047,
      "epoch": 3.435232220609579,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003849178997475856,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21219601184129716,
      "num_tokens": 78578386.0,
      "step": 37870
    },
    {
      "entropy": 5.412988710403442,
      "epoch": 3.4356857764876634,
      "grad_norm": 1.0,
      "learning_rate": 0.00038488966600218365,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20740234702825547,
      "num_tokens": 78588855.0,
      "step": 37875
    },
    {
      "entropy": 5.469479894638061,
      "epoch": 3.4361393323657476,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038486142998423535,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20586646050214769,
      "num_tokens": 78599573.0,
      "step": 37880
    },
    {
      "entropy": 5.577449750900269,
      "epoch": 3.4365928882438315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003848331916943248,
      "loss": 5.1826,
      "mean_token_accuracy": 0.18971046507358552,
      "num_tokens": 78609793.0,
      "step": 37885
    },
    {
      "entropy": 5.43119854927063,
      "epoch": 3.4370464441219157,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003848049511330358,
      "loss": 5.0548,
      "mean_token_accuracy": 0.20207471549510955,
      "num_tokens": 78619637.0,
      "step": 37890
    },
    {
      "entropy": 5.5415435314178465,
      "epoch": 3.4375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003847767083009525,
      "loss": 5.1602,
      "mean_token_accuracy": 0.20228863954544068,
      "num_tokens": 78629889.0,
      "step": 37895
    },
    {
      "entropy": 5.469722652435303,
      "epoch": 3.4379535558780843,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038474846319865876,
      "loss": 5.0479,
      "mean_token_accuracy": 0.20528991818428038,
      "num_tokens": 78639853.0,
      "step": 37900
    },
    {
      "entropy": 5.485693645477295,
      "epoch": 3.4384071117561685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003847202158267389,
      "loss": 5.0385,
      "mean_token_accuracy": 0.2076851323246956,
      "num_tokens": 78649732.0,
      "step": 37905
    },
    {
      "entropy": 5.382607936859131,
      "epoch": 3.4388606676342524,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038469196618577694,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20286835283041,
      "num_tokens": 78659794.0,
      "step": 37910
    },
    {
      "entropy": 5.588103485107422,
      "epoch": 3.4393142235123366,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003846637142763573,
      "loss": 5.2256,
      "mean_token_accuracy": 0.19291912913322448,
      "num_tokens": 78670526.0,
      "step": 37915
    },
    {
      "entropy": 5.494877004623413,
      "epoch": 3.439767779390421,
      "grad_norm": 1.0625,
      "learning_rate": 0.000384635460099064,
      "loss": 5.0798,
      "mean_token_accuracy": 0.20496048927307128,
      "num_tokens": 78680177.0,
      "step": 37920
    },
    {
      "entropy": 5.550471448898316,
      "epoch": 3.440221335268505,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038460720365448154,
      "loss": 5.2661,
      "mean_token_accuracy": 0.19017109423875808,
      "num_tokens": 78690958.0,
      "step": 37925
    },
    {
      "entropy": 5.555388069152832,
      "epoch": 3.4406748911465894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003845789449431941,
      "loss": 5.1319,
      "mean_token_accuracy": 0.19605033695697785,
      "num_tokens": 78700904.0,
      "step": 37930
    },
    {
      "entropy": 5.595401811599731,
      "epoch": 3.4411284470246732,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003845506839657862,
      "loss": 5.2023,
      "mean_token_accuracy": 0.19349896013736725,
      "num_tokens": 78712085.0,
      "step": 37935
    },
    {
      "entropy": 5.500177097320557,
      "epoch": 3.4415820029027575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845224207228422,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20756566375494004,
      "num_tokens": 78722161.0,
      "step": 37940
    },
    {
      "entropy": 5.539780855178833,
      "epoch": 3.4420355587808418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038449415521494667,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19358640909194946,
      "num_tokens": 78732485.0,
      "step": 37945
    },
    {
      "entropy": 5.476840209960938,
      "epoch": 3.442489114658926,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003844658874426841,
      "loss": 5.1187,
      "mean_token_accuracy": 0.19762298315763474,
      "num_tokens": 78744564.0,
      "step": 37950
    },
    {
      "entropy": 5.5162880420684814,
      "epoch": 3.4429426705370103,
      "grad_norm": 1.0625,
      "learning_rate": 0.000384437617406639,
      "loss": 5.0922,
      "mean_token_accuracy": 0.20141702741384507,
      "num_tokens": 78755227.0,
      "step": 37955
    },
    {
      "entropy": 5.540314817428589,
      "epoch": 3.4433962264150946,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003844093451073962,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19578560143709184,
      "num_tokens": 78765393.0,
      "step": 37960
    },
    {
      "entropy": 5.494316053390503,
      "epoch": 3.4438497822931784,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038438107054554027,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19921741932630538,
      "num_tokens": 78775587.0,
      "step": 37965
    },
    {
      "entropy": 5.454352188110351,
      "epoch": 3.4443033381712627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003843527937216558,
      "loss": 5.0182,
      "mean_token_accuracy": 0.21062173843383789,
      "num_tokens": 78785098.0,
      "step": 37970
    },
    {
      "entropy": 5.4635303020477295,
      "epoch": 3.444756894049347,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003843245146363278,
      "loss": 5.0604,
      "mean_token_accuracy": 0.20460571795701982,
      "num_tokens": 78795348.0,
      "step": 37975
    },
    {
      "entropy": 5.588199949264526,
      "epoch": 3.445210449927431,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038429623329014096,
      "loss": 5.2053,
      "mean_token_accuracy": 0.191717292368412,
      "num_tokens": 78805348.0,
      "step": 37980
    },
    {
      "entropy": 5.454150438308716,
      "epoch": 3.445664005805515,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038426794968368015,
      "loss": 5.0219,
      "mean_token_accuracy": 0.2012816607952118,
      "num_tokens": 78814765.0,
      "step": 37985
    },
    {
      "entropy": 5.483701181411743,
      "epoch": 3.4461175616835993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003842396638175303,
      "loss": 5.0566,
      "mean_token_accuracy": 0.2062778189778328,
      "num_tokens": 78824745.0,
      "step": 37990
    },
    {
      "entropy": 5.422488641738892,
      "epoch": 3.4465711175616836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003842113756922763,
      "loss": 5.1689,
      "mean_token_accuracy": 0.19475708454847335,
      "num_tokens": 78835101.0,
      "step": 37995
    },
    {
      "entropy": 5.637726926803589,
      "epoch": 3.447024673439768,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038418308530850333,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18960495591163634,
      "num_tokens": 78845881.0,
      "step": 38000
    },
    {
      "entropy": 5.559360599517822,
      "epoch": 3.447478229317852,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038415479266679624,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19392331093549728,
      "num_tokens": 78856536.0,
      "step": 38005
    },
    {
      "entropy": 5.48722653388977,
      "epoch": 3.4479317851959363,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003841264977677403,
      "loss": 5.0911,
      "mean_token_accuracy": 0.1990675523877144,
      "num_tokens": 78866887.0,
      "step": 38010
    },
    {
      "entropy": 5.48890175819397,
      "epoch": 3.44838534107402,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038409820061192045,
      "loss": 5.0849,
      "mean_token_accuracy": 0.20700473189353943,
      "num_tokens": 78877032.0,
      "step": 38015
    },
    {
      "entropy": 5.451743698120117,
      "epoch": 3.4488388969521044,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038406990119992215,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20529117584228515,
      "num_tokens": 78887237.0,
      "step": 38020
    },
    {
      "entropy": 5.375350904464722,
      "epoch": 3.4492924528301887,
      "grad_norm": 1.125,
      "learning_rate": 0.00038404159953233047,
      "loss": 5.019,
      "mean_token_accuracy": 0.21394042372703553,
      "num_tokens": 78897436.0,
      "step": 38025
    },
    {
      "entropy": 5.525835609436035,
      "epoch": 3.449746008708273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003840132956097307,
      "loss": 5.101,
      "mean_token_accuracy": 0.19723298400640488,
      "num_tokens": 78907691.0,
      "step": 38030
    },
    {
      "entropy": 5.546904182434082,
      "epoch": 3.450199564586357,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038398498943270816,
      "loss": 5.1373,
      "mean_token_accuracy": 0.20043532103300093,
      "num_tokens": 78917517.0,
      "step": 38035
    },
    {
      "entropy": 5.570178127288818,
      "epoch": 3.450653120464441,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003839566810018483,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19953316003084182,
      "num_tokens": 78927835.0,
      "step": 38040
    },
    {
      "entropy": 5.492095470428467,
      "epoch": 3.4511066763425253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038392837031773656,
      "loss": 5.0329,
      "mean_token_accuracy": 0.19985686093568802,
      "num_tokens": 78938853.0,
      "step": 38045
    },
    {
      "entropy": 5.431567716598511,
      "epoch": 3.4515602322206096,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003839000573809583,
      "loss": 5.1,
      "mean_token_accuracy": 0.20206145644187928,
      "num_tokens": 78948508.0,
      "step": 38050
    },
    {
      "entropy": 5.547009992599487,
      "epoch": 3.452013788098694,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038387174219209916,
      "loss": 5.1073,
      "mean_token_accuracy": 0.20160080790519713,
      "num_tokens": 78959002.0,
      "step": 38055
    },
    {
      "entropy": 5.519224548339844,
      "epoch": 3.452467343976778,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003838434247517446,
      "loss": 5.0906,
      "mean_token_accuracy": 0.2005490705370903,
      "num_tokens": 78970358.0,
      "step": 38060
    },
    {
      "entropy": 5.467424821853638,
      "epoch": 3.452920899854862,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003838151050604803,
      "loss": 5.0076,
      "mean_token_accuracy": 0.2110515207052231,
      "num_tokens": 78979408.0,
      "step": 38065
    },
    {
      "entropy": 5.398849725723267,
      "epoch": 3.453374455732946,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038378678311889184,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20757144093513488,
      "num_tokens": 78990411.0,
      "step": 38070
    },
    {
      "entropy": 5.533368873596191,
      "epoch": 3.4538280116110305,
      "grad_norm": 1.046875,
      "learning_rate": 0.000383758458927565,
      "loss": 5.168,
      "mean_token_accuracy": 0.20332372039556504,
      "num_tokens": 79001734.0,
      "step": 38075
    },
    {
      "entropy": 5.5484922409057615,
      "epoch": 3.4542815674891147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003837301324870856,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2070010185241699,
      "num_tokens": 79011381.0,
      "step": 38080
    },
    {
      "entropy": 5.475626993179321,
      "epoch": 3.454735123367199,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038370180379803927,
      "loss": 5.171,
      "mean_token_accuracy": 0.1952717423439026,
      "num_tokens": 79021124.0,
      "step": 38085
    },
    {
      "entropy": 5.566329622268677,
      "epoch": 3.455188679245283,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038367347286101197,
      "loss": 5.2162,
      "mean_token_accuracy": 0.18404396027326583,
      "num_tokens": 79033105.0,
      "step": 38090
    },
    {
      "entropy": 5.538495349884033,
      "epoch": 3.455642235123367,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003836451396765896,
      "loss": 5.1244,
      "mean_token_accuracy": 0.19606753587722778,
      "num_tokens": 79043458.0,
      "step": 38095
    },
    {
      "entropy": 5.516564655303955,
      "epoch": 3.4560957910014514,
      "grad_norm": 0.921875,
      "learning_rate": 0.00038361680424535795,
      "loss": 5.1226,
      "mean_token_accuracy": 0.20226995050907134,
      "num_tokens": 79055186.0,
      "step": 38100
    },
    {
      "entropy": 5.520973634719849,
      "epoch": 3.4565493468795356,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003835884665679032,
      "loss": 5.1613,
      "mean_token_accuracy": 0.19114979803562165,
      "num_tokens": 79065743.0,
      "step": 38105
    },
    {
      "entropy": 5.530795955657959,
      "epoch": 3.45700290275762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003835601266448112,
      "loss": 5.2086,
      "mean_token_accuracy": 0.19055496007204056,
      "num_tokens": 79077285.0,
      "step": 38110
    },
    {
      "entropy": 5.537842559814453,
      "epoch": 3.457456458635704,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003835317844766682,
      "loss": 5.0913,
      "mean_token_accuracy": 0.20022435039281844,
      "num_tokens": 79087885.0,
      "step": 38115
    },
    {
      "entropy": 5.571041822433472,
      "epoch": 3.457910014513788,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038350344006406016,
      "loss": 5.138,
      "mean_token_accuracy": 0.2049359053373337,
      "num_tokens": 79097907.0,
      "step": 38120
    },
    {
      "entropy": 5.596839094161988,
      "epoch": 3.4583635703918723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038347509340757343,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19410427510738373,
      "num_tokens": 79108478.0,
      "step": 38125
    },
    {
      "entropy": 5.5571270942687985,
      "epoch": 3.4588171262699565,
      "grad_norm": 1.25,
      "learning_rate": 0.0003834467445077941,
      "loss": 5.1327,
      "mean_token_accuracy": 0.19701615422964097,
      "num_tokens": 79117988.0,
      "step": 38130
    },
    {
      "entropy": 5.578922891616822,
      "epoch": 3.459270682148041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003834183933653084,
      "loss": 5.119,
      "mean_token_accuracy": 0.20586021840572358,
      "num_tokens": 79127786.0,
      "step": 38135
    },
    {
      "entropy": 5.497344970703125,
      "epoch": 3.4597242380261246,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003833900399807027,
      "loss": 5.084,
      "mean_token_accuracy": 0.21021199524402617,
      "num_tokens": 79137775.0,
      "step": 38140
    },
    {
      "entropy": 5.412124681472778,
      "epoch": 3.460177793904209,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038336168435456345,
      "loss": 5.0056,
      "mean_token_accuracy": 0.20560232698917388,
      "num_tokens": 79147803.0,
      "step": 38145
    },
    {
      "entropy": 5.453021144866943,
      "epoch": 3.460631349782293,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038333332648747684,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20503109246492385,
      "num_tokens": 79158377.0,
      "step": 38150
    },
    {
      "entropy": 5.504535245895386,
      "epoch": 3.4610849056603774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003833049663800295,
      "loss": 5.0987,
      "mean_token_accuracy": 0.20059153884649278,
      "num_tokens": 79168514.0,
      "step": 38155
    },
    {
      "entropy": 5.528931951522827,
      "epoch": 3.4615384615384617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003832766040328078,
      "loss": 5.2111,
      "mean_token_accuracy": 0.19349247813224793,
      "num_tokens": 79177588.0,
      "step": 38160
    },
    {
      "entropy": 5.4706628799438475,
      "epoch": 3.461992017416546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003832482394463984,
      "loss": 5.0627,
      "mean_token_accuracy": 0.20464122146368027,
      "num_tokens": 79188128.0,
      "step": 38165
    },
    {
      "entropy": 5.4211647510528564,
      "epoch": 3.4624455732946298,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003832198726213878,
      "loss": 5.0404,
      "mean_token_accuracy": 0.20047823637723922,
      "num_tokens": 79198928.0,
      "step": 38170
    },
    {
      "entropy": 5.476132392883301,
      "epoch": 3.462899129172714,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038319150355836266,
      "loss": 5.018,
      "mean_token_accuracy": 0.21086929440498353,
      "num_tokens": 79209751.0,
      "step": 38175
    },
    {
      "entropy": 5.506204080581665,
      "epoch": 3.4633526850507983,
      "grad_norm": 1.125,
      "learning_rate": 0.00038316313225790965,
      "loss": 5.1348,
      "mean_token_accuracy": 0.19563247859477997,
      "num_tokens": 79220210.0,
      "step": 38180
    },
    {
      "entropy": 5.510004377365112,
      "epoch": 3.4638062409288826,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003831347587206155,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1981208398938179,
      "num_tokens": 79231530.0,
      "step": 38185
    },
    {
      "entropy": 5.450869798660278,
      "epoch": 3.4642597968069664,
      "grad_norm": 1.1875,
      "learning_rate": 0.000383106382947067,
      "loss": 5.0706,
      "mean_token_accuracy": 0.20153585076332092,
      "num_tokens": 79242771.0,
      "step": 38190
    },
    {
      "entropy": 5.517802619934082,
      "epoch": 3.4647133526850507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038307800493785093,
      "loss": 5.0691,
      "mean_token_accuracy": 0.20399437993764877,
      "num_tokens": 79253713.0,
      "step": 38195
    },
    {
      "entropy": 5.50032868385315,
      "epoch": 3.465166908563135,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038304962469355415,
      "loss": 5.1222,
      "mean_token_accuracy": 0.19908243417739868,
      "num_tokens": 79264087.0,
      "step": 38200
    },
    {
      "entropy": 5.519377088546753,
      "epoch": 3.465620464441219,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003830212422147637,
      "loss": 5.103,
      "mean_token_accuracy": 0.20284634977579116,
      "num_tokens": 79274859.0,
      "step": 38205
    },
    {
      "entropy": 5.519988775253296,
      "epoch": 3.4660740203193035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003829928575020663,
      "loss": 5.1251,
      "mean_token_accuracy": 0.20441804081201553,
      "num_tokens": 79284778.0,
      "step": 38210
    },
    {
      "entropy": 5.5247071266174315,
      "epoch": 3.4665275761973877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038296447055604916,
      "loss": 5.128,
      "mean_token_accuracy": 0.1910233423113823,
      "num_tokens": 79295034.0,
      "step": 38215
    },
    {
      "entropy": 5.426458787918091,
      "epoch": 3.4669811320754715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003829360813772992,
      "loss": 5.0488,
      "mean_token_accuracy": 0.20870323926210405,
      "num_tokens": 79306013.0,
      "step": 38220
    },
    {
      "entropy": 5.517907238006591,
      "epoch": 3.467434687953556,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038290768996640366,
      "loss": 5.1745,
      "mean_token_accuracy": 0.1956264331936836,
      "num_tokens": 79317590.0,
      "step": 38225
    },
    {
      "entropy": 5.588072729110718,
      "epoch": 3.46788824383164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038287929632394957,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19494784325361253,
      "num_tokens": 79328669.0,
      "step": 38230
    },
    {
      "entropy": 5.415354442596436,
      "epoch": 3.4683417997097243,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003828509004505241,
      "loss": 5.0096,
      "mean_token_accuracy": 0.21511538922786713,
      "num_tokens": 79339336.0,
      "step": 38235
    },
    {
      "entropy": 5.499950218200683,
      "epoch": 3.4687953555878086,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003828225023467146,
      "loss": 5.1586,
      "mean_token_accuracy": 0.19754296839237212,
      "num_tokens": 79350199.0,
      "step": 38240
    },
    {
      "entropy": 5.526788949966431,
      "epoch": 3.4692489114658924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003827941020131082,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20777512341737747,
      "num_tokens": 79359906.0,
      "step": 38245
    },
    {
      "entropy": 5.491537809371948,
      "epoch": 3.4697024673439767,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038276569945029225,
      "loss": 5.0082,
      "mean_token_accuracy": 0.19869572222232817,
      "num_tokens": 79371395.0,
      "step": 38250
    },
    {
      "entropy": 5.471735143661499,
      "epoch": 3.470156023222061,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003827372946588543,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20337913781404496,
      "num_tokens": 79381808.0,
      "step": 38255
    },
    {
      "entropy": 5.446611595153809,
      "epoch": 3.4706095791001452,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003827088876393816,
      "loss": 5.0692,
      "mean_token_accuracy": 0.2054697811603546,
      "num_tokens": 79391906.0,
      "step": 38260
    },
    {
      "entropy": 5.468192529678345,
      "epoch": 3.4710631349782295,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038268047839246165,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2052384227514267,
      "num_tokens": 79402522.0,
      "step": 38265
    },
    {
      "entropy": 5.440789651870728,
      "epoch": 3.4715166908563133,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038265206691868195,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20294683128595353,
      "num_tokens": 79412751.0,
      "step": 38270
    },
    {
      "entropy": 5.538848257064819,
      "epoch": 3.4719702467343976,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003826236532186301,
      "loss": 5.2103,
      "mean_token_accuracy": 0.195342817902565,
      "num_tokens": 79424512.0,
      "step": 38275
    },
    {
      "entropy": 5.657304096221924,
      "epoch": 3.472423802612482,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003825952372928936,
      "loss": 5.2574,
      "mean_token_accuracy": 0.19033921211957933,
      "num_tokens": 79434082.0,
      "step": 38280
    },
    {
      "entropy": 5.580122995376587,
      "epoch": 3.472877358490566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038256681914206026,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20013109892606734,
      "num_tokens": 79444469.0,
      "step": 38285
    },
    {
      "entropy": 5.411320209503174,
      "epoch": 3.4733309143686504,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038253839876671765,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20014920830726624,
      "num_tokens": 79455036.0,
      "step": 38290
    },
    {
      "entropy": 5.49574646949768,
      "epoch": 3.473784470246734,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003825099761674535,
      "loss": 5.0893,
      "mean_token_accuracy": 0.20077770203351974,
      "num_tokens": 79464600.0,
      "step": 38295
    },
    {
      "entropy": 5.499926948547364,
      "epoch": 3.4742380261248185,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003824815513448558,
      "loss": 5.112,
      "mean_token_accuracy": 0.20048681646585464,
      "num_tokens": 79474912.0,
      "step": 38300
    },
    {
      "entropy": 5.3578400135040285,
      "epoch": 3.4746915820029027,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038245312429951214,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2058709055185318,
      "num_tokens": 79485283.0,
      "step": 38305
    },
    {
      "entropy": 5.41523323059082,
      "epoch": 3.475145137880987,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003824246950320105,
      "loss": 4.9615,
      "mean_token_accuracy": 0.2016408234834671,
      "num_tokens": 79495323.0,
      "step": 38310
    },
    {
      "entropy": 5.512080144882202,
      "epoch": 3.4755986937590713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003823962635429388,
      "loss": 5.1278,
      "mean_token_accuracy": 0.1987543836236,
      "num_tokens": 79505144.0,
      "step": 38315
    },
    {
      "entropy": 5.564778852462768,
      "epoch": 3.4760522496371555,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003823678298328849,
      "loss": 5.1405,
      "mean_token_accuracy": 0.19917321503162383,
      "num_tokens": 79515402.0,
      "step": 38320
    },
    {
      "entropy": 5.469444131851196,
      "epoch": 3.4765058055152394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038233939390243697,
      "loss": 5.029,
      "mean_token_accuracy": 0.21818387657403945,
      "num_tokens": 79525528.0,
      "step": 38325
    },
    {
      "entropy": 5.4772826671600345,
      "epoch": 3.4769593613933236,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038231095575218303,
      "loss": 5.093,
      "mean_token_accuracy": 0.20458054393529893,
      "num_tokens": 79535991.0,
      "step": 38330
    },
    {
      "entropy": 5.568944644927979,
      "epoch": 3.477412917271408,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003822825153827112,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19389541149139405,
      "num_tokens": 79545890.0,
      "step": 38335
    },
    {
      "entropy": 5.542179298400879,
      "epoch": 3.477866473149492,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038225407279460964,
      "loss": 5.1662,
      "mean_token_accuracy": 0.2013234257698059,
      "num_tokens": 79556363.0,
      "step": 38340
    },
    {
      "entropy": 5.495380449295044,
      "epoch": 3.478320029027576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003822256279884665,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1984474316239357,
      "num_tokens": 79566924.0,
      "step": 38345
    },
    {
      "entropy": 5.467992067337036,
      "epoch": 3.4787735849056602,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038219718096487007,
      "loss": 5.0917,
      "mean_token_accuracy": 0.20218331664800643,
      "num_tokens": 79577322.0,
      "step": 38350
    },
    {
      "entropy": 5.5861917495727536,
      "epoch": 3.4792271407837445,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038216873172440855,
      "loss": 5.2078,
      "mean_token_accuracy": 0.19283760339021683,
      "num_tokens": 79588436.0,
      "step": 38355
    },
    {
      "entropy": 5.584131479263306,
      "epoch": 3.4796806966618288,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003821402802676704,
      "loss": 5.1161,
      "mean_token_accuracy": 0.19914246946573258,
      "num_tokens": 79599015.0,
      "step": 38360
    },
    {
      "entropy": 5.479602241516114,
      "epoch": 3.480134252539913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003821118265952439,
      "loss": 5.0675,
      "mean_token_accuracy": 0.2079191952943802,
      "num_tokens": 79608443.0,
      "step": 38365
    },
    {
      "entropy": 5.4841924667358395,
      "epoch": 3.4805878084179973,
      "grad_norm": 1.125,
      "learning_rate": 0.0003820833707077176,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2015484094619751,
      "num_tokens": 79619010.0,
      "step": 38370
    },
    {
      "entropy": 5.545584726333618,
      "epoch": 3.481041364296081,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003820549126056799,
      "loss": 5.1032,
      "mean_token_accuracy": 0.2011138379573822,
      "num_tokens": 79629665.0,
      "step": 38375
    },
    {
      "entropy": 5.387294673919678,
      "epoch": 3.4814949201741654,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003820264522897193,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20652102828025817,
      "num_tokens": 79640578.0,
      "step": 38380
    },
    {
      "entropy": 5.542365217208863,
      "epoch": 3.4819484760522497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003819979897604243,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19907448589801788,
      "num_tokens": 79649576.0,
      "step": 38385
    },
    {
      "entropy": 5.502372074127197,
      "epoch": 3.482402031930334,
      "grad_norm": 1.0,
      "learning_rate": 0.00038196952501838376,
      "loss": 5.111,
      "mean_token_accuracy": 0.19292163252830505,
      "num_tokens": 79659926.0,
      "step": 38390
    },
    {
      "entropy": 5.521496295928955,
      "epoch": 3.4828555878084178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038194105806418617,
      "loss": 5.233,
      "mean_token_accuracy": 0.19263335168361664,
      "num_tokens": 79669969.0,
      "step": 38395
    },
    {
      "entropy": 5.518358945846558,
      "epoch": 3.483309143686502,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003819125888984202,
      "loss": 5.0391,
      "mean_token_accuracy": 0.20349562913179398,
      "num_tokens": 79679706.0,
      "step": 38400
    },
    {
      "entropy": 5.497051763534546,
      "epoch": 3.4837626995645863,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038188411752167466,
      "loss": 5.0604,
      "mean_token_accuracy": 0.20241684168577195,
      "num_tokens": 79691543.0,
      "step": 38405
    },
    {
      "entropy": 5.527233028411866,
      "epoch": 3.4842162554426706,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003818556439345383,
      "loss": 5.1176,
      "mean_token_accuracy": 0.20081815719604493,
      "num_tokens": 79701657.0,
      "step": 38410
    },
    {
      "entropy": 5.469538688659668,
      "epoch": 3.484669811320755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038182716813760004,
      "loss": 5.1104,
      "mean_token_accuracy": 0.2022862732410431,
      "num_tokens": 79711615.0,
      "step": 38415
    },
    {
      "entropy": 5.503113317489624,
      "epoch": 3.485123367198839,
      "grad_norm": 1.0,
      "learning_rate": 0.0003817986901314487,
      "loss": 5.1162,
      "mean_token_accuracy": 0.2072195827960968,
      "num_tokens": 79721427.0,
      "step": 38420
    },
    {
      "entropy": 5.514422130584717,
      "epoch": 3.485576923076923,
      "grad_norm": 1.375,
      "learning_rate": 0.0003817702099166732,
      "loss": 5.1221,
      "mean_token_accuracy": 0.20154304206371307,
      "num_tokens": 79732565.0,
      "step": 38425
    },
    {
      "entropy": 5.436441802978516,
      "epoch": 3.486030478955007,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0003817417274938626,
      "loss": 4.9392,
      "mean_token_accuracy": 0.21143908351659774,
      "num_tokens": 79744496.0,
      "step": 38430
    },
    {
      "entropy": 5.5141724109649655,
      "epoch": 3.4864840348330914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003817132428636058,
      "loss": 5.1769,
      "mean_token_accuracy": 0.19680997133255004,
      "num_tokens": 79755626.0,
      "step": 38435
    },
    {
      "entropy": 5.489849424362182,
      "epoch": 3.4869375907111757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038168475602649203,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19640839248895645,
      "num_tokens": 79765447.0,
      "step": 38440
    },
    {
      "entropy": 5.491004133224488,
      "epoch": 3.48739114658926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003816562669831103,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19680186361074448,
      "num_tokens": 79776047.0,
      "step": 38445
    },
    {
      "entropy": 5.496452903747558,
      "epoch": 3.487844702467344,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038162777573404975,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20356062799692154,
      "num_tokens": 79786061.0,
      "step": 38450
    },
    {
      "entropy": 5.466193580627442,
      "epoch": 3.488298258345428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038159928227989964,
      "loss": 5.1221,
      "mean_token_accuracy": 0.2032811239361763,
      "num_tokens": 79796844.0,
      "step": 38455
    },
    {
      "entropy": 5.5125490665435795,
      "epoch": 3.4887518142235123,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003815707866212493,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20572724491357802,
      "num_tokens": 79808120.0,
      "step": 38460
    },
    {
      "entropy": 5.503214740753174,
      "epoch": 3.4892053701015966,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003815422887586878,
      "loss": 5.1046,
      "mean_token_accuracy": 0.20401752293109893,
      "num_tokens": 79818901.0,
      "step": 38465
    },
    {
      "entropy": 5.499232196807862,
      "epoch": 3.489658925979681,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003815137886928047,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19549148678779601,
      "num_tokens": 79829090.0,
      "step": 38470
    },
    {
      "entropy": 5.5773945331573485,
      "epoch": 3.490112481857765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003814852864241892,
      "loss": 5.1333,
      "mean_token_accuracy": 0.19610553830862046,
      "num_tokens": 79838751.0,
      "step": 38475
    },
    {
      "entropy": 5.504512357711792,
      "epoch": 3.490566037735849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038145678195343095,
      "loss": 5.1819,
      "mean_token_accuracy": 0.1991170421242714,
      "num_tokens": 79849410.0,
      "step": 38480
    },
    {
      "entropy": 5.594820976257324,
      "epoch": 3.491019593613933,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003814282752811193,
      "loss": 5.1366,
      "mean_token_accuracy": 0.19722556620836257,
      "num_tokens": 79859279.0,
      "step": 38485
    },
    {
      "entropy": 5.6127434253692625,
      "epoch": 3.4914731494920175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003813997664078438,
      "loss": 5.0764,
      "mean_token_accuracy": 0.2005778208374977,
      "num_tokens": 79869535.0,
      "step": 38490
    },
    {
      "entropy": 5.609490013122558,
      "epoch": 3.4919267053701017,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038137125533419403,
      "loss": 5.2459,
      "mean_token_accuracy": 0.18312567323446274,
      "num_tokens": 79881396.0,
      "step": 38495
    },
    {
      "entropy": 5.436487627029419,
      "epoch": 3.4923802612481856,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003813427420607596,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19868526309728624,
      "num_tokens": 79891250.0,
      "step": 38500
    },
    {
      "entropy": 5.444912099838257,
      "epoch": 3.49283381712627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038131422658813004,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19937701672315597,
      "num_tokens": 79901982.0,
      "step": 38505
    },
    {
      "entropy": 5.5058599472045895,
      "epoch": 3.493287373004354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038128570891689527,
      "loss": 5.1607,
      "mean_token_accuracy": 0.1975599154829979,
      "num_tokens": 79912480.0,
      "step": 38510
    },
    {
      "entropy": 5.545873928070068,
      "epoch": 3.4937409288824384,
      "grad_norm": 1.09375,
      "learning_rate": 0.000381257189047645,
      "loss": 5.1204,
      "mean_token_accuracy": 0.20434805005788803,
      "num_tokens": 79923793.0,
      "step": 38515
    },
    {
      "entropy": 5.516493892669677,
      "epoch": 3.4941944847605226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038122866698096894,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19979458153247834,
      "num_tokens": 79933858.0,
      "step": 38520
    },
    {
      "entropy": 5.516898918151855,
      "epoch": 3.494648040638607,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038120014271745693,
      "loss": 5.1163,
      "mean_token_accuracy": 0.19977129399776458,
      "num_tokens": 79944093.0,
      "step": 38525
    },
    {
      "entropy": 5.52666163444519,
      "epoch": 3.4951015965166907,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000381171616257699,
      "loss": 5.1341,
      "mean_token_accuracy": 0.19755065441131592,
      "num_tokens": 79955072.0,
      "step": 38530
    },
    {
      "entropy": 5.540650320053101,
      "epoch": 3.495555152394775,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038114308760228493,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19918352961540223,
      "num_tokens": 79966949.0,
      "step": 38535
    },
    {
      "entropy": 5.455707120895386,
      "epoch": 3.4960087082728593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003811145567518048,
      "loss": 5.0358,
      "mean_token_accuracy": 0.19881874769926072,
      "num_tokens": 79977109.0,
      "step": 38540
    },
    {
      "entropy": 5.536735343933105,
      "epoch": 3.4964622641509435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038108602370684854,
      "loss": 5.132,
      "mean_token_accuracy": 0.19582446664571762,
      "num_tokens": 79987094.0,
      "step": 38545
    },
    {
      "entropy": 5.511499071121216,
      "epoch": 3.4969158200290273,
      "grad_norm": 1.5,
      "learning_rate": 0.0003810574884680063,
      "loss": 5.2029,
      "mean_token_accuracy": 0.20695100724697113,
      "num_tokens": 79997503.0,
      "step": 38550
    },
    {
      "entropy": 5.508684301376343,
      "epoch": 3.4973693759071116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038102895103586817,
      "loss": 5.0861,
      "mean_token_accuracy": 0.20216822177171706,
      "num_tokens": 80007745.0,
      "step": 38555
    },
    {
      "entropy": 5.559760904312133,
      "epoch": 3.497822931785196,
      "grad_norm": 1.734375,
      "learning_rate": 0.0003810004114110243,
      "loss": 5.1521,
      "mean_token_accuracy": 0.19413113594055176,
      "num_tokens": 80019330.0,
      "step": 38560
    },
    {
      "entropy": 5.518109607696533,
      "epoch": 3.49827648766328,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003809718695940649,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19838087856769562,
      "num_tokens": 80029732.0,
      "step": 38565
    },
    {
      "entropy": 5.459735870361328,
      "epoch": 3.4987300435413644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003809433255855803,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20417468696832658,
      "num_tokens": 80041052.0,
      "step": 38570
    },
    {
      "entropy": 5.506283712387085,
      "epoch": 3.4991835994194487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038091477938616075,
      "loss": 5.1252,
      "mean_token_accuracy": 0.20250172764062882,
      "num_tokens": 80051929.0,
      "step": 38575
    },
    {
      "entropy": 5.539838075637817,
      "epoch": 3.4996371552975325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038088623099639653,
      "loss": 5.1734,
      "mean_token_accuracy": 0.19560556411743163,
      "num_tokens": 80061153.0,
      "step": 38580
    },
    {
      "entropy": 5.51757025718689,
      "epoch": 3.5000907111756168,
      "grad_norm": 1.03125,
      "learning_rate": 0.000380857680416878,
      "loss": 5.073,
      "mean_token_accuracy": 0.20513146966695786,
      "num_tokens": 80070560.0,
      "step": 38585
    },
    {
      "entropy": 5.660133266448975,
      "epoch": 3.500544267053701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003808291276481957,
      "loss": 5.166,
      "mean_token_accuracy": 0.19729899764060974,
      "num_tokens": 80081141.0,
      "step": 38590
    },
    {
      "entropy": 5.432094717025757,
      "epoch": 3.5009978229317853,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003808005726909401,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21274574249982833,
      "num_tokens": 80092114.0,
      "step": 38595
    },
    {
      "entropy": 5.4440850734710695,
      "epoch": 3.501451378809869,
      "grad_norm": 1.0,
      "learning_rate": 0.00038077201554570175,
      "loss": 5.0197,
      "mean_token_accuracy": 0.2088201180100441,
      "num_tokens": 80101966.0,
      "step": 38600
    },
    {
      "entropy": 5.611114549636841,
      "epoch": 3.5019049346879534,
      "grad_norm": 1.78125,
      "learning_rate": 0.0003807434562130711,
      "loss": 5.2155,
      "mean_token_accuracy": 0.20229257494211197,
      "num_tokens": 80113420.0,
      "step": 38605
    },
    {
      "entropy": 5.495820808410644,
      "epoch": 3.5023584905660377,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003807148946936387,
      "loss": 5.1079,
      "mean_token_accuracy": 0.19650898873806,
      "num_tokens": 80123398.0,
      "step": 38610
    },
    {
      "entropy": 5.490578126907349,
      "epoch": 3.502812046444122,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003806863309879956,
      "loss": 5.0423,
      "mean_token_accuracy": 0.21046415567398072,
      "num_tokens": 80133885.0,
      "step": 38615
    },
    {
      "entropy": 5.515842437744141,
      "epoch": 3.503265602322206,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003806577650967321,
      "loss": 5.0178,
      "mean_token_accuracy": 0.21454178243875505,
      "num_tokens": 80144168.0,
      "step": 38620
    },
    {
      "entropy": 5.4646519184112545,
      "epoch": 3.5037191582002905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003806291970204391,
      "loss": 5.0757,
      "mean_token_accuracy": 0.2017328530550003,
      "num_tokens": 80153956.0,
      "step": 38625
    },
    {
      "entropy": 5.51147780418396,
      "epoch": 3.5041727140783747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038060062675970743,
      "loss": 5.1266,
      "mean_token_accuracy": 0.19237859547138214,
      "num_tokens": 80163617.0,
      "step": 38630
    },
    {
      "entropy": 5.461778068542481,
      "epoch": 3.5046262699564585,
      "grad_norm": 1.0,
      "learning_rate": 0.0003805720543151278,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20705001950263976,
      "num_tokens": 80174049.0,
      "step": 38635
    },
    {
      "entropy": 5.550410985946655,
      "epoch": 3.505079825834543,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003805434796872912,
      "loss": 5.1226,
      "mean_token_accuracy": 0.20249897092580796,
      "num_tokens": 80185043.0,
      "step": 38640
    },
    {
      "entropy": 5.4191131591796875,
      "epoch": 3.505533381712627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038051490287678856,
      "loss": 5.062,
      "mean_token_accuracy": 0.2022160768508911,
      "num_tokens": 80194810.0,
      "step": 38645
    },
    {
      "entropy": 5.472354888916016,
      "epoch": 3.5059869375907113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003804863238842109,
      "loss": 5.0834,
      "mean_token_accuracy": 0.2057754099369049,
      "num_tokens": 80204861.0,
      "step": 38650
    },
    {
      "entropy": 5.510281324386597,
      "epoch": 3.506440493468795,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003804577427101491,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20768103897571563,
      "num_tokens": 80214447.0,
      "step": 38655
    },
    {
      "entropy": 5.437661600112915,
      "epoch": 3.5068940493468794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003804291593551944,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20814015716314316,
      "num_tokens": 80224213.0,
      "step": 38660
    },
    {
      "entropy": 5.4721763134002686,
      "epoch": 3.5073476052249637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003804005738199377,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20494537204504013,
      "num_tokens": 80234161.0,
      "step": 38665
    },
    {
      "entropy": 5.527026081085205,
      "epoch": 3.507801161103048,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038037198610497037,
      "loss": 5.0765,
      "mean_token_accuracy": 0.20242204815149306,
      "num_tokens": 80243788.0,
      "step": 38670
    },
    {
      "entropy": 5.516359758377075,
      "epoch": 3.5082547169811322,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003803433962108835,
      "loss": 5.0879,
      "mean_token_accuracy": 0.20387008637189866,
      "num_tokens": 80254658.0,
      "step": 38675
    },
    {
      "entropy": 5.4985206604003904,
      "epoch": 3.5087082728592165,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003803148041382683,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19795295149087905,
      "num_tokens": 80264820.0,
      "step": 38680
    },
    {
      "entropy": 5.435080909729004,
      "epoch": 3.5091618287373003,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038028620988771617,
      "loss": 5.0101,
      "mean_token_accuracy": 0.2088076278567314,
      "num_tokens": 80274639.0,
      "step": 38685
    },
    {
      "entropy": 5.4854559898376465,
      "epoch": 3.5096153846153846,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003802576134598183,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20589190423488618,
      "num_tokens": 80284542.0,
      "step": 38690
    },
    {
      "entropy": 5.467952394485474,
      "epoch": 3.510068940493469,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038022901485516625,
      "loss": 5.0335,
      "mean_token_accuracy": 0.20488673448562622,
      "num_tokens": 80294405.0,
      "step": 38695
    },
    {
      "entropy": 5.492648410797119,
      "epoch": 3.510522496371553,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038020041407435133,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19741743355989455,
      "num_tokens": 80305326.0,
      "step": 38700
    },
    {
      "entropy": 5.466943597793579,
      "epoch": 3.510976052249637,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038017181111796506,
      "loss": 5.0945,
      "mean_token_accuracy": 0.20291339755058288,
      "num_tokens": 80315423.0,
      "step": 38705
    },
    {
      "entropy": 5.49110164642334,
      "epoch": 3.511429608127721,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003801432059865989,
      "loss": 5.1271,
      "mean_token_accuracy": 0.19768763184547425,
      "num_tokens": 80325372.0,
      "step": 38710
    },
    {
      "entropy": 5.56231803894043,
      "epoch": 3.5118831640058055,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038011459868084446,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20022729337215422,
      "num_tokens": 80335467.0,
      "step": 38715
    },
    {
      "entropy": 5.45820803642273,
      "epoch": 3.5123367198838897,
      "grad_norm": 1.0,
      "learning_rate": 0.0003800859892012933,
      "loss": 5.0759,
      "mean_token_accuracy": 0.20893123596906663,
      "num_tokens": 80345420.0,
      "step": 38720
    },
    {
      "entropy": 5.425785875320434,
      "epoch": 3.512790275761974,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003800573775485372,
      "loss": 5.0727,
      "mean_token_accuracy": 0.21112728863954544,
      "num_tokens": 80354369.0,
      "step": 38725
    },
    {
      "entropy": 5.5228517055511475,
      "epoch": 3.5132438316400583,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038002876372316774,
      "loss": 5.156,
      "mean_token_accuracy": 0.20183427929878234,
      "num_tokens": 80365292.0,
      "step": 38730
    },
    {
      "entropy": 5.56667308807373,
      "epoch": 3.513697387518142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038000014772577657,
      "loss": 5.1258,
      "mean_token_accuracy": 0.20054859220981597,
      "num_tokens": 80376340.0,
      "step": 38735
    },
    {
      "entropy": 5.453342342376709,
      "epoch": 3.5141509433962264,
      "grad_norm": 1.0,
      "learning_rate": 0.0003799715295569557,
      "loss": 5.0551,
      "mean_token_accuracy": 0.2015736147761345,
      "num_tokens": 80386193.0,
      "step": 38740
    },
    {
      "entropy": 5.630128526687622,
      "epoch": 3.5146044992743106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003799429092172968,
      "loss": 5.2988,
      "mean_token_accuracy": 0.19105596095323563,
      "num_tokens": 80397055.0,
      "step": 38745
    },
    {
      "entropy": 5.526120996475219,
      "epoch": 3.515058055152395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003799142867073918,
      "loss": 5.1495,
      "mean_token_accuracy": 0.19381249845027923,
      "num_tokens": 80407375.0,
      "step": 38750
    },
    {
      "entropy": 5.541679430007934,
      "epoch": 3.5155116110304787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037988566202783256,
      "loss": 5.1331,
      "mean_token_accuracy": 0.19740869700908661,
      "num_tokens": 80417942.0,
      "step": 38755
    },
    {
      "entropy": 5.436580467224121,
      "epoch": 3.515965166908563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003798570351792112,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2071322277188301,
      "num_tokens": 80428716.0,
      "step": 38760
    },
    {
      "entropy": 5.490742301940918,
      "epoch": 3.5164187227866472,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037982840616211964,
      "loss": 5.1468,
      "mean_token_accuracy": 0.2017776131629944,
      "num_tokens": 80439688.0,
      "step": 38765
    },
    {
      "entropy": 5.553225898742676,
      "epoch": 3.5168722786647315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037979977497714984,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19425411224365235,
      "num_tokens": 80450434.0,
      "step": 38770
    },
    {
      "entropy": 5.530705785751342,
      "epoch": 3.5173258345428158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037977114162489404,
      "loss": 5.1828,
      "mean_token_accuracy": 0.19472991079092025,
      "num_tokens": 80462187.0,
      "step": 38775
    },
    {
      "entropy": 5.490103387832642,
      "epoch": 3.5177793904209,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003797425061059443,
      "loss": 5.0376,
      "mean_token_accuracy": 0.19908739924430846,
      "num_tokens": 80473049.0,
      "step": 38780
    },
    {
      "entropy": 5.472080326080322,
      "epoch": 3.5182329462989843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037971386842089287,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20976944714784623,
      "num_tokens": 80484119.0,
      "step": 38785
    },
    {
      "entropy": 5.465299272537232,
      "epoch": 3.518686502177068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003796852285703319,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20882650464773178,
      "num_tokens": 80494882.0,
      "step": 38790
    },
    {
      "entropy": 5.521879243850708,
      "epoch": 3.5191400580551524,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003796565865548538,
      "loss": 5.1373,
      "mean_token_accuracy": 0.20234149396419526,
      "num_tokens": 80505362.0,
      "step": 38795
    },
    {
      "entropy": 5.510329723358154,
      "epoch": 3.5195936139332367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003796279423750508,
      "loss": 5.1603,
      "mean_token_accuracy": 0.19536410570144652,
      "num_tokens": 80516027.0,
      "step": 38800
    },
    {
      "entropy": 5.547585201263428,
      "epoch": 3.5200471698113205,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003795992960315153,
      "loss": 5.129,
      "mean_token_accuracy": 0.20036587119102478,
      "num_tokens": 80527045.0,
      "step": 38805
    },
    {
      "entropy": 5.474673271179199,
      "epoch": 3.5205007256894048,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037957064752483966,
      "loss": 5.0291,
      "mean_token_accuracy": 0.2107683852314949,
      "num_tokens": 80537373.0,
      "step": 38810
    },
    {
      "entropy": 5.437787771224976,
      "epoch": 3.520954281567489,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003795419968556164,
      "loss": 5.0556,
      "mean_token_accuracy": 0.20082564204931258,
      "num_tokens": 80546632.0,
      "step": 38815
    },
    {
      "entropy": 5.4690868854522705,
      "epoch": 3.5214078374455733,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003795133440244381,
      "loss": 5.0277,
      "mean_token_accuracy": 0.2068807676434517,
      "num_tokens": 80556694.0,
      "step": 38820
    },
    {
      "entropy": 5.535464811325073,
      "epoch": 3.5218613933236576,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003794846890318971,
      "loss": 5.1659,
      "mean_token_accuracy": 0.19441474825143815,
      "num_tokens": 80567834.0,
      "step": 38825
    },
    {
      "entropy": 5.53923659324646,
      "epoch": 3.522314949201742,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003794560318785861,
      "loss": 5.0839,
      "mean_token_accuracy": 0.2015620321035385,
      "num_tokens": 80578299.0,
      "step": 38830
    },
    {
      "entropy": 5.506132173538208,
      "epoch": 3.522768505079826,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037942737256509786,
      "loss": 5.0925,
      "mean_token_accuracy": 0.19845218509435653,
      "num_tokens": 80588620.0,
      "step": 38835
    },
    {
      "entropy": 5.619432401657105,
      "epoch": 3.52322206095791,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003793987110920248,
      "loss": 5.3157,
      "mean_token_accuracy": 0.18697492629289628,
      "num_tokens": 80599076.0,
      "step": 38840
    },
    {
      "entropy": 5.608066082000732,
      "epoch": 3.523675616835994,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037937004745995987,
      "loss": 5.2155,
      "mean_token_accuracy": 0.1935548886656761,
      "num_tokens": 80609440.0,
      "step": 38845
    },
    {
      "entropy": 5.598013687133789,
      "epoch": 3.5241291727140784,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003793413816694957,
      "loss": 5.1612,
      "mean_token_accuracy": 0.19569020122289657,
      "num_tokens": 80619169.0,
      "step": 38850
    },
    {
      "entropy": 5.483826732635498,
      "epoch": 3.5245827285921627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003793127137212252,
      "loss": 5.022,
      "mean_token_accuracy": 0.2066024050116539,
      "num_tokens": 80628719.0,
      "step": 38855
    },
    {
      "entropy": 5.49820990562439,
      "epoch": 3.5250362844702465,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037928404361574125,
      "loss": 5.0617,
      "mean_token_accuracy": 0.1952075332403183,
      "num_tokens": 80638391.0,
      "step": 38860
    },
    {
      "entropy": 5.4685760021209715,
      "epoch": 3.525489840348331,
      "grad_norm": 1.125,
      "learning_rate": 0.0003792553713536367,
      "loss": 5.1453,
      "mean_token_accuracy": 0.19791561216115952,
      "num_tokens": 80647720.0,
      "step": 38865
    },
    {
      "entropy": 5.421838712692261,
      "epoch": 3.525943396226415,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003792266969355044,
      "loss": 5.0606,
      "mean_token_accuracy": 0.20489675104618071,
      "num_tokens": 80658841.0,
      "step": 38870
    },
    {
      "entropy": 5.5432336807250975,
      "epoch": 3.5263969521044993,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037919802036193753,
      "loss": 5.1172,
      "mean_token_accuracy": 0.19847540110349654,
      "num_tokens": 80668735.0,
      "step": 38875
    },
    {
      "entropy": 5.5647704124450685,
      "epoch": 3.5268505079825836,
      "grad_norm": 2.234375,
      "learning_rate": 0.00037916934163352903,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20355530381202697,
      "num_tokens": 80680757.0,
      "step": 38880
    },
    {
      "entropy": 5.464684629440308,
      "epoch": 3.527304063860668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000379140660750872,
      "loss": 5.019,
      "mean_token_accuracy": 0.2030344381928444,
      "num_tokens": 80689833.0,
      "step": 38885
    },
    {
      "entropy": 5.471528005599976,
      "epoch": 3.5277576197387517,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003791119777145595,
      "loss": 5.0137,
      "mean_token_accuracy": 0.2055925115942955,
      "num_tokens": 80700347.0,
      "step": 38890
    },
    {
      "entropy": 5.4108446598052975,
      "epoch": 3.528211175616836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003790832925251849,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21821944266557694,
      "num_tokens": 80710559.0,
      "step": 38895
    },
    {
      "entropy": 5.444446086883545,
      "epoch": 3.52866473149492,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003790546051833412,
      "loss": 5.1225,
      "mean_token_accuracy": 0.2019718438386917,
      "num_tokens": 80720805.0,
      "step": 38900
    },
    {
      "entropy": 5.551635885238648,
      "epoch": 3.5291182873730045,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003790259156896217,
      "loss": 5.1224,
      "mean_token_accuracy": 0.1865704521536827,
      "num_tokens": 80730937.0,
      "step": 38905
    },
    {
      "entropy": 5.548355436325073,
      "epoch": 3.5295718432510883,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037899722404461983,
      "loss": 5.1085,
      "mean_token_accuracy": 0.20301583260297776,
      "num_tokens": 80741597.0,
      "step": 38910
    },
    {
      "entropy": 5.51317868232727,
      "epoch": 3.5300253991291726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003789685302489289,
      "loss": 5.089,
      "mean_token_accuracy": 0.19522259682416915,
      "num_tokens": 80752496.0,
      "step": 38915
    },
    {
      "entropy": 5.491311168670654,
      "epoch": 3.530478955007257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037893983430314226,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20440713316202164,
      "num_tokens": 80762329.0,
      "step": 38920
    },
    {
      "entropy": 5.476008367538452,
      "epoch": 3.530932510885341,
      "grad_norm": 1.9375,
      "learning_rate": 0.00037891113620785327,
      "loss": 5.1789,
      "mean_token_accuracy": 0.19701503664255143,
      "num_tokens": 80773152.0,
      "step": 38925
    },
    {
      "entropy": 5.471571683883667,
      "epoch": 3.5313860667634254,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037888243596365557,
      "loss": 5.0653,
      "mean_token_accuracy": 0.2034836456179619,
      "num_tokens": 80783502.0,
      "step": 38930
    },
    {
      "entropy": 5.505853366851807,
      "epoch": 3.5318396226415096,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037885373357114265,
      "loss": 4.9992,
      "mean_token_accuracy": 0.20489597618579863,
      "num_tokens": 80792644.0,
      "step": 38935
    },
    {
      "entropy": 5.521279335021973,
      "epoch": 3.5322931785195935,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000378825029030908,
      "loss": 5.1477,
      "mean_token_accuracy": 0.1981002688407898,
      "num_tokens": 80803184.0,
      "step": 38940
    },
    {
      "entropy": 5.518962240219116,
      "epoch": 3.5327467343976777,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003787963223435453,
      "loss": 5.0865,
      "mean_token_accuracy": 0.1953973203897476,
      "num_tokens": 80813353.0,
      "step": 38945
    },
    {
      "entropy": 5.526396322250366,
      "epoch": 3.533200290275762,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003787676135096482,
      "loss": 5.1883,
      "mean_token_accuracy": 0.19445238262414932,
      "num_tokens": 80824663.0,
      "step": 38950
    },
    {
      "entropy": 5.612483930587769,
      "epoch": 3.5336538461538463,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003787389025298104,
      "loss": 5.2001,
      "mean_token_accuracy": 0.19100759625434877,
      "num_tokens": 80835507.0,
      "step": 38955
    },
    {
      "entropy": 5.478664302825928,
      "epoch": 3.53410740203193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003787101894046257,
      "loss": 5.0328,
      "mean_token_accuracy": 0.2080032393336296,
      "num_tokens": 80845035.0,
      "step": 38960
    },
    {
      "entropy": 5.479308032989502,
      "epoch": 3.5345609579100143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037868147413468785,
      "loss": 5.0317,
      "mean_token_accuracy": 0.2070591628551483,
      "num_tokens": 80854992.0,
      "step": 38965
    },
    {
      "entropy": 5.548110675811768,
      "epoch": 3.5350145137880986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003786527567205907,
      "loss": 5.1713,
      "mean_token_accuracy": 0.1883214920759201,
      "num_tokens": 80864658.0,
      "step": 38970
    },
    {
      "entropy": 5.560680723190307,
      "epoch": 3.535468069666183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037862403716292805,
      "loss": 5.0723,
      "mean_token_accuracy": 0.203813336789608,
      "num_tokens": 80873964.0,
      "step": 38975
    },
    {
      "entropy": 5.502188634872437,
      "epoch": 3.535921625544267,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037859531546229405,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19746956825256348,
      "num_tokens": 80883752.0,
      "step": 38980
    },
    {
      "entropy": 5.498976135253907,
      "epoch": 3.5363751814223514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003785665916192824,
      "loss": 5.1345,
      "mean_token_accuracy": 0.19851363003253936,
      "num_tokens": 80894265.0,
      "step": 38985
    },
    {
      "entropy": 5.49329981803894,
      "epoch": 3.5368287373004357,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037853786563448724,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2093917801976204,
      "num_tokens": 80904246.0,
      "step": 38990
    },
    {
      "entropy": 5.513451147079468,
      "epoch": 3.5372822931785195,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003785091375085027,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19802480936050415,
      "num_tokens": 80913595.0,
      "step": 38995
    },
    {
      "entropy": 5.428896045684814,
      "epoch": 3.5377358490566038,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003784804072419228,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19931127727031708,
      "num_tokens": 80924754.0,
      "step": 39000
    },
    {
      "epoch": 3.5377358490566038,
      "eval_entropy": 5.241348891883954,
      "eval_loss": 5.171875,
      "eval_mean_token_accuracy": 0.20536979815050352,
      "eval_num_tokens": 80924754.0,
      "eval_runtime": 26.7416,
      "eval_samples_per_second": 1322.282,
      "eval_steps_per_second": 165.285,
      "step": 39000
    },
    {
      "entropy": 5.457676124572754,
      "epoch": 3.538189404934688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003784516748353417,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20649310797452927,
      "num_tokens": 80935318.0,
      "step": 39005
    },
    {
      "entropy": 5.501255989074707,
      "epoch": 3.5386429608127723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003784229402893536,
      "loss": 5.1359,
      "mean_token_accuracy": 0.20131127089262008,
      "num_tokens": 80944886.0,
      "step": 39010
    },
    {
      "entropy": 5.541685438156128,
      "epoch": 3.539096516690856,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003783942036045528,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19780416190624237,
      "num_tokens": 80955566.0,
      "step": 39015
    },
    {
      "entropy": 5.5507200241088865,
      "epoch": 3.5395500725689404,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037836546478153344,
      "loss": 5.1729,
      "mean_token_accuracy": 0.19942402690649033,
      "num_tokens": 80965285.0,
      "step": 39020
    },
    {
      "entropy": 5.432683897018433,
      "epoch": 3.5400036284470247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003783367238208899,
      "loss": 5.0731,
      "mean_token_accuracy": 0.20443056523799896,
      "num_tokens": 80976402.0,
      "step": 39025
    },
    {
      "entropy": 5.588996171951294,
      "epoch": 3.540457184325109,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037830798072321665,
      "loss": 5.1953,
      "mean_token_accuracy": 0.19463563114404678,
      "num_tokens": 80988408.0,
      "step": 39030
    },
    {
      "entropy": 5.52737979888916,
      "epoch": 3.540910740203193,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037827923548910806,
      "loss": 5.0862,
      "mean_token_accuracy": 0.20245834290981293,
      "num_tokens": 80999099.0,
      "step": 39035
    },
    {
      "entropy": 5.4547521591186525,
      "epoch": 3.5413642960812775,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037825048811915854,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21202886402606963,
      "num_tokens": 81009350.0,
      "step": 39040
    },
    {
      "entropy": 5.454828119277954,
      "epoch": 3.5418178519593613,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037822173861396256,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20707014054059983,
      "num_tokens": 81019851.0,
      "step": 39045
    },
    {
      "entropy": 5.449585771560669,
      "epoch": 3.5422714078374455,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037819298697411475,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1977041929960251,
      "num_tokens": 81028987.0,
      "step": 39050
    },
    {
      "entropy": 5.461740255355835,
      "epoch": 3.54272496371553,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037816423320020964,
      "loss": 5.0299,
      "mean_token_accuracy": 0.2079747200012207,
      "num_tokens": 81038081.0,
      "step": 39055
    },
    {
      "entropy": 5.4913355827331545,
      "epoch": 3.543178519593614,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037813547729284195,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20126013904809953,
      "num_tokens": 81048066.0,
      "step": 39060
    },
    {
      "entropy": 5.327743864059448,
      "epoch": 3.543632075471698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037810671925260633,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21836926490068437,
      "num_tokens": 81057421.0,
      "step": 39065
    },
    {
      "entropy": 5.428357553482056,
      "epoch": 3.544085631349782,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037807795908009745,
      "loss": 5.0759,
      "mean_token_accuracy": 0.20146837830543518,
      "num_tokens": 81068435.0,
      "step": 39070
    },
    {
      "entropy": 5.584997177124023,
      "epoch": 3.5445391872278664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037804919677591006,
      "loss": 5.1223,
      "mean_token_accuracy": 0.1998835101723671,
      "num_tokens": 81079378.0,
      "step": 39075
    },
    {
      "entropy": 5.520350599288941,
      "epoch": 3.5449927431059507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003780204323406391,
      "loss": 5.1029,
      "mean_token_accuracy": 0.19882228821516038,
      "num_tokens": 81089872.0,
      "step": 39080
    },
    {
      "entropy": 5.455804824829102,
      "epoch": 3.545446298984035,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003779916657748793,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20612133294343948,
      "num_tokens": 81099653.0,
      "step": 39085
    },
    {
      "entropy": 5.432977485656738,
      "epoch": 3.5458998548621192,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003779628970792256,
      "loss": 5.0759,
      "mean_token_accuracy": 0.20597352981567382,
      "num_tokens": 81109840.0,
      "step": 39090
    },
    {
      "entropy": 5.4578088283538815,
      "epoch": 3.546353410740203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000377934126254273,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20780884772539138,
      "num_tokens": 81120991.0,
      "step": 39095
    },
    {
      "entropy": 5.574812698364258,
      "epoch": 3.5468069666182873,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037790535330061635,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20400545597076417,
      "num_tokens": 81129831.0,
      "step": 39100
    },
    {
      "entropy": 5.491153573989868,
      "epoch": 3.5472605224963716,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003778765782188508,
      "loss": 5.0854,
      "mean_token_accuracy": 0.207486891746521,
      "num_tokens": 81140604.0,
      "step": 39105
    },
    {
      "entropy": 5.489930248260498,
      "epoch": 3.547714078374456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037784780100957145,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20753582417964936,
      "num_tokens": 81150591.0,
      "step": 39110
    },
    {
      "entropy": 5.570986080169678,
      "epoch": 3.5481676342525397,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0003778190216733733,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1923343285918236,
      "num_tokens": 81161444.0,
      "step": 39115
    },
    {
      "entropy": 5.545912981033325,
      "epoch": 3.548621190130624,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037779024021085167,
      "loss": 5.0213,
      "mean_token_accuracy": 0.21048613786697387,
      "num_tokens": 81171216.0,
      "step": 39120
    },
    {
      "entropy": 5.444218635559082,
      "epoch": 3.549074746008708,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003777614566226016,
      "loss": 5.1297,
      "mean_token_accuracy": 0.2032440185546875,
      "num_tokens": 81181665.0,
      "step": 39125
    },
    {
      "entropy": 5.419114255905152,
      "epoch": 3.5495283018867925,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003777326709092184,
      "loss": 5.0393,
      "mean_token_accuracy": 0.2049001008272171,
      "num_tokens": 81191683.0,
      "step": 39130
    },
    {
      "entropy": 5.55358510017395,
      "epoch": 3.5499818577648767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037770388307129747,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20221709311008454,
      "num_tokens": 81201211.0,
      "step": 39135
    },
    {
      "entropy": 5.472090244293213,
      "epoch": 3.550435413642961,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037767509310943407,
      "loss": 5.0453,
      "mean_token_accuracy": 0.2042008951306343,
      "num_tokens": 81210668.0,
      "step": 39140
    },
    {
      "entropy": 5.440426778793335,
      "epoch": 3.5508889695210453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003776463010242235,
      "loss": 4.951,
      "mean_token_accuracy": 0.21195433139801026,
      "num_tokens": 81221073.0,
      "step": 39145
    },
    {
      "entropy": 5.468371915817261,
      "epoch": 3.551342525399129,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037761750681626134,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19665692746639252,
      "num_tokens": 81231687.0,
      "step": 39150
    },
    {
      "entropy": 5.528838014602661,
      "epoch": 3.5517960812772134,
      "grad_norm": 1.0,
      "learning_rate": 0.00037758871048614295,
      "loss": 5.1383,
      "mean_token_accuracy": 0.1969103619456291,
      "num_tokens": 81242323.0,
      "step": 39155
    },
    {
      "entropy": 5.468768692016601,
      "epoch": 3.5522496371552976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003775599120344638,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21598308831453322,
      "num_tokens": 81252543.0,
      "step": 39160
    },
    {
      "entropy": 5.512844467163086,
      "epoch": 3.5527031930333814,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037753111146181966,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19588735997676848,
      "num_tokens": 81262180.0,
      "step": 39165
    },
    {
      "entropy": 5.557901763916016,
      "epoch": 3.5531567489114657,
      "grad_norm": 1.140625,
      "learning_rate": 0.000377502308768806,
      "loss": 5.2721,
      "mean_token_accuracy": 0.18494407832622528,
      "num_tokens": 81272088.0,
      "step": 39170
    },
    {
      "entropy": 5.490322589874268,
      "epoch": 3.55361030478955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003774735039560184,
      "loss": 5.047,
      "mean_token_accuracy": 0.2072245106101036,
      "num_tokens": 81281723.0,
      "step": 39175
    },
    {
      "entropy": 5.577476119995117,
      "epoch": 3.5540638606676342,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037744469702405273,
      "loss": 5.1492,
      "mean_token_accuracy": 0.20338619202375413,
      "num_tokens": 81292030.0,
      "step": 39180
    },
    {
      "entropy": 5.487811470031739,
      "epoch": 3.5545174165457185,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003774158879735045,
      "loss": 5.0479,
      "mean_token_accuracy": 0.2087007388472557,
      "num_tokens": 81302283.0,
      "step": 39185
    },
    {
      "entropy": 5.5238487243652346,
      "epoch": 3.5549709724238028,
      "grad_norm": 1.125,
      "learning_rate": 0.0003773870768049697,
      "loss": 5.1249,
      "mean_token_accuracy": 0.19750234931707383,
      "num_tokens": 81313058.0,
      "step": 39190
    },
    {
      "entropy": 5.480618953704834,
      "epoch": 3.555424528301887,
      "grad_norm": 1.21875,
      "learning_rate": 0.000377358263519044,
      "loss": 5.0633,
      "mean_token_accuracy": 0.2053656354546547,
      "num_tokens": 81323827.0,
      "step": 39195
    },
    {
      "entropy": 5.504893016815186,
      "epoch": 3.555878084179971,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003773294481163234,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19940701127052307,
      "num_tokens": 81333106.0,
      "step": 39200
    },
    {
      "entropy": 5.487491512298584,
      "epoch": 3.556331640058055,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003773006305974037,
      "loss": 5.0646,
      "mean_token_accuracy": 0.20367009937763214,
      "num_tokens": 81343245.0,
      "step": 39205
    },
    {
      "entropy": 5.507078742980957,
      "epoch": 3.5567851959361394,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037727181096288077,
      "loss": 5.0674,
      "mean_token_accuracy": 0.20868695825338363,
      "num_tokens": 81352917.0,
      "step": 39210
    },
    {
      "entropy": 5.435956573486328,
      "epoch": 3.5572387518142237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003772429892133509,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2056937411427498,
      "num_tokens": 81362838.0,
      "step": 39215
    },
    {
      "entropy": 5.481378173828125,
      "epoch": 3.5576923076923075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003772141653494099,
      "loss": 5.0061,
      "mean_token_accuracy": 0.206903900206089,
      "num_tokens": 81373395.0,
      "step": 39220
    },
    {
      "entropy": 5.4636729717254635,
      "epoch": 3.5581458635703918,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003771853393716538,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2059513583779335,
      "num_tokens": 81383976.0,
      "step": 39225
    },
    {
      "entropy": 5.45892825126648,
      "epoch": 3.558599419448476,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000377156511280679,
      "loss": 5.0547,
      "mean_token_accuracy": 0.1977224290370941,
      "num_tokens": 81394328.0,
      "step": 39230
    },
    {
      "entropy": 5.500485754013061,
      "epoch": 3.5590529753265603,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003771276810770815,
      "loss": 5.0646,
      "mean_token_accuracy": 0.20032123923301698,
      "num_tokens": 81405451.0,
      "step": 39235
    },
    {
      "entropy": 5.531237697601318,
      "epoch": 3.5595065312046446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003770988487614575,
      "loss": 5.1547,
      "mean_token_accuracy": 0.19412773698568345,
      "num_tokens": 81416589.0,
      "step": 39240
    },
    {
      "entropy": 5.460300159454346,
      "epoch": 3.559960087082729,
      "grad_norm": 3.671875,
      "learning_rate": 0.0003770700143344033,
      "loss": 5.0505,
      "mean_token_accuracy": 0.20624420195817947,
      "num_tokens": 81428553.0,
      "step": 39245
    },
    {
      "entropy": 5.625598382949829,
      "epoch": 3.5604136429608126,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037704117779651514,
      "loss": 5.2139,
      "mean_token_accuracy": 0.19187749177217484,
      "num_tokens": 81438720.0,
      "step": 39250
    },
    {
      "entropy": 5.4491219997406,
      "epoch": 3.560867198838897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037701233914838947,
      "loss": 5.1264,
      "mean_token_accuracy": 0.20051170140504837,
      "num_tokens": 81448142.0,
      "step": 39255
    },
    {
      "entropy": 5.5200447082519535,
      "epoch": 3.561320754716981,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003769834983906226,
      "loss": 5.1382,
      "mean_token_accuracy": 0.19990457594394684,
      "num_tokens": 81459448.0,
      "step": 39260
    },
    {
      "entropy": 5.512991809844971,
      "epoch": 3.5617743105950654,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037695465552381105,
      "loss": 5.0641,
      "mean_token_accuracy": 0.2066972956061363,
      "num_tokens": 81470356.0,
      "step": 39265
    },
    {
      "entropy": 5.4804760932922365,
      "epoch": 3.5622278664731493,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037692581054855125,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20333431512117386,
      "num_tokens": 81480472.0,
      "step": 39270
    },
    {
      "entropy": 5.441479301452636,
      "epoch": 3.5626814223512335,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037689696346543964,
      "loss": 5.0168,
      "mean_token_accuracy": 0.21209034621715545,
      "num_tokens": 81489949.0,
      "step": 39275
    },
    {
      "entropy": 5.3989622592926025,
      "epoch": 3.563134978229318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037686811427507284,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2055161252617836,
      "num_tokens": 81500276.0,
      "step": 39280
    },
    {
      "entropy": 5.468477249145508,
      "epoch": 3.563588534107402,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037683926297804756,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20158110111951827,
      "num_tokens": 81510431.0,
      "step": 39285
    },
    {
      "entropy": 5.472375774383545,
      "epoch": 3.5640420899854863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003768104095749603,
      "loss": 5.0537,
      "mean_token_accuracy": 0.1992087781429291,
      "num_tokens": 81520718.0,
      "step": 39290
    },
    {
      "entropy": 5.490348529815674,
      "epoch": 3.5644956458635706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003767815540664079,
      "loss": 5.1477,
      "mean_token_accuracy": 0.2009011834859848,
      "num_tokens": 81531904.0,
      "step": 39295
    },
    {
      "entropy": 5.494182062149048,
      "epoch": 3.5649492017416544,
      "grad_norm": 1.0,
      "learning_rate": 0.00037675269645298697,
      "loss": 5.0755,
      "mean_token_accuracy": 0.19644154608249664,
      "num_tokens": 81542093.0,
      "step": 39300
    },
    {
      "entropy": 5.466160869598388,
      "epoch": 3.5654027576197387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003767238367352944,
      "loss": 4.9809,
      "mean_token_accuracy": 0.211318439245224,
      "num_tokens": 81552020.0,
      "step": 39305
    },
    {
      "entropy": 5.418948078155518,
      "epoch": 3.565856313497823,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037669497491392683,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21283842772245407,
      "num_tokens": 81561644.0,
      "step": 39310
    },
    {
      "entropy": 5.513252782821655,
      "epoch": 3.566309869375907,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037666611098948144,
      "loss": 5.1024,
      "mean_token_accuracy": 0.20082972347736358,
      "num_tokens": 81572158.0,
      "step": 39315
    },
    {
      "entropy": 5.553066062927246,
      "epoch": 3.566763425253991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003766372449625549,
      "loss": 5.1642,
      "mean_token_accuracy": 0.1901743084192276,
      "num_tokens": 81582335.0,
      "step": 39320
    },
    {
      "entropy": 5.617348146438599,
      "epoch": 3.5672169811320753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003766083768337442,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1978563115000725,
      "num_tokens": 81593004.0,
      "step": 39325
    },
    {
      "entropy": 5.5308167934417725,
      "epoch": 3.5676705370101596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003765795066036464,
      "loss": 5.0678,
      "mean_token_accuracy": 0.20423521250486373,
      "num_tokens": 81603656.0,
      "step": 39330
    },
    {
      "entropy": 5.440824174880982,
      "epoch": 3.568124092888244,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037655063427285854,
      "loss": 5.0898,
      "mean_token_accuracy": 0.20190368294715882,
      "num_tokens": 81613039.0,
      "step": 39335
    },
    {
      "entropy": 5.462597942352295,
      "epoch": 3.568577648766328,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037652175984197764,
      "loss": 5.0475,
      "mean_token_accuracy": 0.2056397870182991,
      "num_tokens": 81624410.0,
      "step": 39340
    },
    {
      "entropy": 5.533057832717896,
      "epoch": 3.5690312046444124,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037649288331160086,
      "loss": 5.0849,
      "mean_token_accuracy": 0.20004383325576783,
      "num_tokens": 81635160.0,
      "step": 39345
    },
    {
      "entropy": 5.507817506790161,
      "epoch": 3.5694847605224966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037646400468232546,
      "loss": 5.0448,
      "mean_token_accuracy": 0.19942005723714828,
      "num_tokens": 81645686.0,
      "step": 39350
    },
    {
      "entropy": 5.4426697254180905,
      "epoch": 3.5699383164005805,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037643512395474853,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20807765871286393,
      "num_tokens": 81656330.0,
      "step": 39355
    },
    {
      "entropy": 5.417167520523071,
      "epoch": 3.5703918722786647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003764062411294674,
      "loss": 5.056,
      "mean_token_accuracy": 0.2004078894853592,
      "num_tokens": 81665895.0,
      "step": 39360
    },
    {
      "entropy": 5.603829574584961,
      "epoch": 3.570845428156749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037637735620707935,
      "loss": 5.1295,
      "mean_token_accuracy": 0.19645078629255294,
      "num_tokens": 81676616.0,
      "step": 39365
    },
    {
      "entropy": 5.542173957824707,
      "epoch": 3.5712989840348333,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003763484691881818,
      "loss": 5.0903,
      "mean_token_accuracy": 0.19555405974388124,
      "num_tokens": 81687169.0,
      "step": 39370
    },
    {
      "entropy": 5.526067924499512,
      "epoch": 3.571752539912917,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037631958007337195,
      "loss": 5.0657,
      "mean_token_accuracy": 0.20386257767677307,
      "num_tokens": 81698291.0,
      "step": 39375
    },
    {
      "entropy": 5.492502355575562,
      "epoch": 3.5722060957910013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003762906888632475,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19821131974458694,
      "num_tokens": 81708750.0,
      "step": 39380
    },
    {
      "entropy": 5.401931619644165,
      "epoch": 3.5726596516690856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003762617955584058,
      "loss": 5.0665,
      "mean_token_accuracy": 0.20423953533172606,
      "num_tokens": 81719474.0,
      "step": 39385
    },
    {
      "entropy": 5.57227463722229,
      "epoch": 3.57311320754717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003762329001594443,
      "loss": 5.1506,
      "mean_token_accuracy": 0.20087490379810333,
      "num_tokens": 81730092.0,
      "step": 39390
    },
    {
      "entropy": 5.530776262283325,
      "epoch": 3.573566763425254,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003762040026669606,
      "loss": 5.0663,
      "mean_token_accuracy": 0.19584383815526962,
      "num_tokens": 81741235.0,
      "step": 39395
    },
    {
      "entropy": 5.46149172782898,
      "epoch": 3.5740203193033384,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003761751030815524,
      "loss": 5.018,
      "mean_token_accuracy": 0.20981938242912293,
      "num_tokens": 81751434.0,
      "step": 39400
    },
    {
      "entropy": 5.509510612487793,
      "epoch": 3.5744738751814222,
      "grad_norm": 1.125,
      "learning_rate": 0.0003761462014038173,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19440154135227203,
      "num_tokens": 81763935.0,
      "step": 39405
    },
    {
      "entropy": 5.5500298023223875,
      "epoch": 3.5749274310595065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003761172976343528,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19818716049194335,
      "num_tokens": 81775007.0,
      "step": 39410
    },
    {
      "entropy": 5.484109592437744,
      "epoch": 3.5753809869375908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037608839177375705,
      "loss": 5.136,
      "mean_token_accuracy": 0.20311056524515153,
      "num_tokens": 81785690.0,
      "step": 39415
    },
    {
      "entropy": 5.435385608673096,
      "epoch": 3.575834542815675,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0003760594838226275,
      "loss": 5.0693,
      "mean_token_accuracy": 0.20513621717691422,
      "num_tokens": 81796143.0,
      "step": 39420
    },
    {
      "entropy": 5.608633995056152,
      "epoch": 3.576288098693759,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003760305737815621,
      "loss": 5.174,
      "mean_token_accuracy": 0.20002432465553283,
      "num_tokens": 81806607.0,
      "step": 39425
    },
    {
      "entropy": 5.472738218307495,
      "epoch": 3.576741654571843,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003760016616511586,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21093593090772628,
      "num_tokens": 81817216.0,
      "step": 39430
    },
    {
      "entropy": 5.457575511932373,
      "epoch": 3.5771952104499274,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037597274743201507,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19823893904685974,
      "num_tokens": 81827211.0,
      "step": 39435
    },
    {
      "entropy": 5.4608314514160154,
      "epoch": 3.5776487663280117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003759438311247294,
      "loss": 5.0237,
      "mean_token_accuracy": 0.21101326495409012,
      "num_tokens": 81837365.0,
      "step": 39440
    },
    {
      "entropy": 5.478324842453003,
      "epoch": 3.578102322206096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003759149127298995,
      "loss": 5.035,
      "mean_token_accuracy": 0.20863971263170242,
      "num_tokens": 81847803.0,
      "step": 39445
    },
    {
      "entropy": 5.457315397262573,
      "epoch": 3.57855587808418,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037588599224812363,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20573839247226716,
      "num_tokens": 81858123.0,
      "step": 39450
    },
    {
      "entropy": 5.395592308044433,
      "epoch": 3.579009433962264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003758570696799997,
      "loss": 5.0315,
      "mean_token_accuracy": 0.21138629019260408,
      "num_tokens": 81868668.0,
      "step": 39455
    },
    {
      "entropy": 5.475000667572021,
      "epoch": 3.5794629898403483,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037582814502612575,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21423047184944152,
      "num_tokens": 81877482.0,
      "step": 39460
    },
    {
      "entropy": 5.431591510772705,
      "epoch": 3.5799165457184325,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003757992182871001,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21536302715539932,
      "num_tokens": 81886700.0,
      "step": 39465
    },
    {
      "entropy": 5.4435042381286625,
      "epoch": 3.580370101596517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037577028946352107,
      "loss": 5.0315,
      "mean_token_accuracy": 0.20673301070928574,
      "num_tokens": 81896446.0,
      "step": 39470
    },
    {
      "entropy": 5.471945953369141,
      "epoch": 3.5808236574746006,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003757413585559866,
      "loss": 5.0721,
      "mean_token_accuracy": 0.20986426323652269,
      "num_tokens": 81907418.0,
      "step": 39475
    },
    {
      "entropy": 5.483234024047851,
      "epoch": 3.581277213352685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037571242556509526,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20820750743150712,
      "num_tokens": 81917940.0,
      "step": 39480
    },
    {
      "entropy": 5.4354691982269285,
      "epoch": 3.581730769230769,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003756834904914453,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20618746876716615,
      "num_tokens": 81928158.0,
      "step": 39485
    },
    {
      "entropy": 5.514724588394165,
      "epoch": 3.5821843251088534,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003756545533356351,
      "loss": 5.0994,
      "mean_token_accuracy": 0.2047171413898468,
      "num_tokens": 81938111.0,
      "step": 39490
    },
    {
      "entropy": 5.51537709236145,
      "epoch": 3.5826378809869377,
      "grad_norm": 0.953125,
      "learning_rate": 0.000375625614098263,
      "loss": 5.0461,
      "mean_token_accuracy": 0.20916946828365326,
      "num_tokens": 81948798.0,
      "step": 39495
    },
    {
      "entropy": 5.525959014892578,
      "epoch": 3.583091436865022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037559667277992765,
      "loss": 5.1299,
      "mean_token_accuracy": 0.20212218165397644,
      "num_tokens": 81959240.0,
      "step": 39500
    },
    {
      "entropy": 5.505724382400513,
      "epoch": 3.5835449927431062,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003755677293812274,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20703878104686738,
      "num_tokens": 81969929.0,
      "step": 39505
    },
    {
      "entropy": 5.536700820922851,
      "epoch": 3.58399854862119,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003755387839027609,
      "loss": 5.1356,
      "mean_token_accuracy": 0.1933029532432556,
      "num_tokens": 81981007.0,
      "step": 39510
    },
    {
      "entropy": 5.545278692245484,
      "epoch": 3.5844521044992743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037550983634512683,
      "loss": 5.2091,
      "mean_token_accuracy": 0.1927129164338112,
      "num_tokens": 81991172.0,
      "step": 39515
    },
    {
      "entropy": 5.452325916290283,
      "epoch": 3.5849056603773586,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003754808867089236,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2142704114317894,
      "num_tokens": 82001282.0,
      "step": 39520
    },
    {
      "entropy": 5.499295949935913,
      "epoch": 3.5853592162554424,
      "grad_norm": 1.0,
      "learning_rate": 0.00037545193499475003,
      "loss": 5.0781,
      "mean_token_accuracy": 0.20106895714998246,
      "num_tokens": 82011929.0,
      "step": 39525
    },
    {
      "entropy": 5.564361476898194,
      "epoch": 3.5858127721335267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003754229812032049,
      "loss": 5.1129,
      "mean_token_accuracy": 0.20369602888822555,
      "num_tokens": 82021663.0,
      "step": 39530
    },
    {
      "entropy": 5.499837207794189,
      "epoch": 3.586266328011611,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037539402533488694,
      "loss": 5.1324,
      "mean_token_accuracy": 0.1966558262705803,
      "num_tokens": 82031547.0,
      "step": 39535
    },
    {
      "entropy": 5.5362693786621096,
      "epoch": 3.586719883889695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003753650673903949,
      "loss": 5.1335,
      "mean_token_accuracy": 0.20317527204751967,
      "num_tokens": 82041722.0,
      "step": 39540
    },
    {
      "entropy": 5.5425803661346436,
      "epoch": 3.5871734397677795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037533610737032773,
      "loss": 5.0942,
      "mean_token_accuracy": 0.20413534045219422,
      "num_tokens": 82052715.0,
      "step": 39545
    },
    {
      "entropy": 5.494831180572509,
      "epoch": 3.5876269956458637,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037530714527528433,
      "loss": 5.0391,
      "mean_token_accuracy": 0.21376466751098633,
      "num_tokens": 82062709.0,
      "step": 39550
    },
    {
      "entropy": 5.515194940567016,
      "epoch": 3.588080551523948,
      "grad_norm": 1.0,
      "learning_rate": 0.00037527818110586354,
      "loss": 5.0838,
      "mean_token_accuracy": 0.2001422092318535,
      "num_tokens": 82072731.0,
      "step": 39555
    },
    {
      "entropy": 5.452774906158448,
      "epoch": 3.588534107402032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003752492148626644,
      "loss": 5.0172,
      "mean_token_accuracy": 0.21125207841396332,
      "num_tokens": 82083216.0,
      "step": 39560
    },
    {
      "entropy": 5.511906719207763,
      "epoch": 3.588987663280116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000375220246546286,
      "loss": 5.1172,
      "mean_token_accuracy": 0.1965319722890854,
      "num_tokens": 82094482.0,
      "step": 39565
    },
    {
      "entropy": 5.503869152069091,
      "epoch": 3.5894412191582004,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003751912761573273,
      "loss": 5.0452,
      "mean_token_accuracy": 0.204571732878685,
      "num_tokens": 82104988.0,
      "step": 39570
    },
    {
      "entropy": 5.455096626281739,
      "epoch": 3.5898947750362846,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037516230369638755,
      "loss": 5.0932,
      "mean_token_accuracy": 0.20045477747917176,
      "num_tokens": 82114455.0,
      "step": 39575
    },
    {
      "entropy": 5.509126806259156,
      "epoch": 3.5903483309143684,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003751333291640658,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19868023693561554,
      "num_tokens": 82125001.0,
      "step": 39580
    },
    {
      "entropy": 5.527381420135498,
      "epoch": 3.5908018867924527,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037510435256096124,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20503968000411987,
      "num_tokens": 82134586.0,
      "step": 39585
    },
    {
      "entropy": 5.4232094287872314,
      "epoch": 3.591255442670537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037507537388767326,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20272277742624284,
      "num_tokens": 82144612.0,
      "step": 39590
    },
    {
      "entropy": 5.439705085754395,
      "epoch": 3.5917089985486212,
      "grad_norm": 1.109375,
      "learning_rate": 0.000375046393144801,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20738562643527986,
      "num_tokens": 82154913.0,
      "step": 39595
    },
    {
      "entropy": 5.518754577636718,
      "epoch": 3.5921625544267055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003750174103329438,
      "loss": 5.12,
      "mean_token_accuracy": 0.20040839910507202,
      "num_tokens": 82164469.0,
      "step": 39600
    },
    {
      "entropy": 5.5335181713104244,
      "epoch": 3.59261611030479,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037498842545270105,
      "loss": 5.0875,
      "mean_token_accuracy": 0.20288585871458054,
      "num_tokens": 82174575.0,
      "step": 39605
    },
    {
      "entropy": 5.5291454792022705,
      "epoch": 3.5930696661828736,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037495943850467226,
      "loss": 5.0547,
      "mean_token_accuracy": 0.20411105155944825,
      "num_tokens": 82184661.0,
      "step": 39610
    },
    {
      "entropy": 5.557375764846801,
      "epoch": 3.593523222060958,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037493044948945677,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19803069680929183,
      "num_tokens": 82194877.0,
      "step": 39615
    },
    {
      "entropy": 5.48955192565918,
      "epoch": 3.593976777939042,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00037490145840765414,
      "loss": 5.079,
      "mean_token_accuracy": 0.2025805562734604,
      "num_tokens": 82206278.0,
      "step": 39620
    },
    {
      "entropy": 5.510198593139648,
      "epoch": 3.5944303338171264,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037487246525986384,
      "loss": 5.0658,
      "mean_token_accuracy": 0.20142273753881454,
      "num_tokens": 82216140.0,
      "step": 39625
    },
    {
      "entropy": 5.552016067504883,
      "epoch": 3.59488388969521,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003748434700466856,
      "loss": 5.1481,
      "mean_token_accuracy": 0.20528950095176696,
      "num_tokens": 82226239.0,
      "step": 39630
    },
    {
      "entropy": 5.46241307258606,
      "epoch": 3.5953374455732945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003748144727687188,
      "loss": 5.0568,
      "mean_token_accuracy": 0.2027061700820923,
      "num_tokens": 82235795.0,
      "step": 39635
    },
    {
      "entropy": 5.460946750640869,
      "epoch": 3.5957910014513788,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037478547342656345,
      "loss": 4.995,
      "mean_token_accuracy": 0.21022588908672332,
      "num_tokens": 82245644.0,
      "step": 39640
    },
    {
      "entropy": 5.490400743484497,
      "epoch": 3.596244557329463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000374756472020819,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2100496247410774,
      "num_tokens": 82255608.0,
      "step": 39645
    },
    {
      "entropy": 5.4860132217407225,
      "epoch": 3.5966981132075473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037472746855208525,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20550766587257385,
      "num_tokens": 82265488.0,
      "step": 39650
    },
    {
      "entropy": 5.503595399856567,
      "epoch": 3.5971516690856316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003746984630209621,
      "loss": 5.0806,
      "mean_token_accuracy": 0.1978297233581543,
      "num_tokens": 82276866.0,
      "step": 39655
    },
    {
      "entropy": 5.466819143295288,
      "epoch": 3.5976052249637154,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037466945542804933,
      "loss": 5.045,
      "mean_token_accuracy": 0.20255929082632065,
      "num_tokens": 82287125.0,
      "step": 39660
    },
    {
      "entropy": 5.547029733657837,
      "epoch": 3.5980587808417996,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037464044577394676,
      "loss": 5.0905,
      "mean_token_accuracy": 0.20133448392152786,
      "num_tokens": 82296887.0,
      "step": 39665
    },
    {
      "entropy": 5.573977661132813,
      "epoch": 3.598512336719884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003746114340592544,
      "loss": 5.1633,
      "mean_token_accuracy": 0.19410266429185868,
      "num_tokens": 82306928.0,
      "step": 39670
    },
    {
      "entropy": 5.4939733982086185,
      "epoch": 3.598965892597968,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037458242028457216,
      "loss": 5.1095,
      "mean_token_accuracy": 0.20486395806074142,
      "num_tokens": 82317873.0,
      "step": 39675
    },
    {
      "entropy": 5.390803575515747,
      "epoch": 3.599419448476052,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037455340445050023,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21367934495210647,
      "num_tokens": 82328184.0,
      "step": 39680
    },
    {
      "entropy": 5.480040979385376,
      "epoch": 3.5998730043541363,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00037452438655763843,
      "loss": 5.0977,
      "mean_token_accuracy": 0.20559158623218537,
      "num_tokens": 82338191.0,
      "step": 39685
    },
    {
      "entropy": 5.547538471221924,
      "epoch": 3.6003265602322205,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037449536660658696,
      "loss": 5.1184,
      "mean_token_accuracy": 0.19895536750555037,
      "num_tokens": 82348095.0,
      "step": 39690
    },
    {
      "entropy": 5.460559129714966,
      "epoch": 3.600780116110305,
      "grad_norm": 1.328125,
      "learning_rate": 0.000374466344597946,
      "loss": 4.9538,
      "mean_token_accuracy": 0.20976660251617432,
      "num_tokens": 82358216.0,
      "step": 39695
    },
    {
      "entropy": 5.378363561630249,
      "epoch": 3.601233671988389,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003744373205323157,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20753444731235504,
      "num_tokens": 82368581.0,
      "step": 39700
    },
    {
      "entropy": 5.484187841415405,
      "epoch": 3.6016872278664733,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003744082944102962,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19838568121194838,
      "num_tokens": 82379609.0,
      "step": 39705
    },
    {
      "entropy": 5.537354803085327,
      "epoch": 3.6021407837445576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003743792662324878,
      "loss": 5.1004,
      "mean_token_accuracy": 0.20151766687631606,
      "num_tokens": 82390589.0,
      "step": 39710
    },
    {
      "entropy": 5.456128740310669,
      "epoch": 3.6025943396226414,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037435023599949086,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20184091329574586,
      "num_tokens": 82400915.0,
      "step": 39715
    },
    {
      "entropy": 5.492248582839966,
      "epoch": 3.6030478955007257,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037432120371190585,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20487271398305892,
      "num_tokens": 82411544.0,
      "step": 39720
    },
    {
      "entropy": 5.428424453735351,
      "epoch": 3.60350145137881,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037429216937033294,
      "loss": 5.0247,
      "mean_token_accuracy": 0.20564854890108109,
      "num_tokens": 82422391.0,
      "step": 39725
    },
    {
      "entropy": 5.452558708190918,
      "epoch": 3.6039550072568938,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037426313297537263,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20417656749486923,
      "num_tokens": 82433564.0,
      "step": 39730
    },
    {
      "entropy": 5.478642511367798,
      "epoch": 3.604408563134978,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003742340945276255,
      "loss": 5.0989,
      "mean_token_accuracy": 0.20133401304483414,
      "num_tokens": 82445005.0,
      "step": 39735
    },
    {
      "entropy": 5.488031244277954,
      "epoch": 3.6048621190130623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000374205054027692,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19766490608453752,
      "num_tokens": 82455998.0,
      "step": 39740
    },
    {
      "entropy": 5.451558065414429,
      "epoch": 3.6053156748911466,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037417601147617265,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20585781037807466,
      "num_tokens": 82466789.0,
      "step": 39745
    },
    {
      "entropy": 5.559834718704224,
      "epoch": 3.605769230769231,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037414696687366816,
      "loss": 5.1938,
      "mean_token_accuracy": 0.195155967772007,
      "num_tokens": 82478181.0,
      "step": 39750
    },
    {
      "entropy": 5.520868015289307,
      "epoch": 3.606222786647315,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037411792022077917,
      "loss": 5.0263,
      "mean_token_accuracy": 0.2039795532822609,
      "num_tokens": 82488452.0,
      "step": 39755
    },
    {
      "entropy": 5.537273693084717,
      "epoch": 3.6066763425253994,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003740888715181063,
      "loss": 5.0922,
      "mean_token_accuracy": 0.20461297929286956,
      "num_tokens": 82498179.0,
      "step": 39760
    },
    {
      "entropy": 5.439065885543823,
      "epoch": 3.607129898403483,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037405982076625026,
      "loss": 5.1104,
      "mean_token_accuracy": 0.20874058157205583,
      "num_tokens": 82508242.0,
      "step": 39765
    },
    {
      "entropy": 5.5782942295074465,
      "epoch": 3.6075834542815675,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037403076796581195,
      "loss": 5.0948,
      "mean_token_accuracy": 0.1979919493198395,
      "num_tokens": 82519363.0,
      "step": 39770
    },
    {
      "entropy": 5.56104884147644,
      "epoch": 3.6080370101596517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003740017131173921,
      "loss": 5.1006,
      "mean_token_accuracy": 0.2011740729212761,
      "num_tokens": 82529745.0,
      "step": 39775
    },
    {
      "entropy": 5.535565280914307,
      "epoch": 3.608490566037736,
      "grad_norm": 1.0,
      "learning_rate": 0.0003739726562215916,
      "loss": 5.1102,
      "mean_token_accuracy": 0.20354532450437546,
      "num_tokens": 82539573.0,
      "step": 39780
    },
    {
      "entropy": 5.559164428710938,
      "epoch": 3.60894412191582,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037394359727901136,
      "loss": 5.1339,
      "mean_token_accuracy": 0.2022797167301178,
      "num_tokens": 82551313.0,
      "step": 39785
    },
    {
      "entropy": 5.479604244232178,
      "epoch": 3.609397677793904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003739145362902523,
      "loss": 5.0445,
      "mean_token_accuracy": 0.20295277684926988,
      "num_tokens": 82561283.0,
      "step": 39790
    },
    {
      "entropy": 5.452059602737426,
      "epoch": 3.6098512336719883,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003738854732559153,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21000578105449677,
      "num_tokens": 82571118.0,
      "step": 39795
    },
    {
      "entropy": 5.478925800323486,
      "epoch": 3.6103047895500726,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003738564081766017,
      "loss": 5.0541,
      "mean_token_accuracy": 0.2052207574248314,
      "num_tokens": 82581663.0,
      "step": 39800
    },
    {
      "entropy": 5.435070896148682,
      "epoch": 3.610758345428157,
      "grad_norm": 1.125,
      "learning_rate": 0.00037382734105291234,
      "loss": 5.0307,
      "mean_token_accuracy": 0.1999860092997551,
      "num_tokens": 82592593.0,
      "step": 39805
    },
    {
      "entropy": 5.509869050979614,
      "epoch": 3.611211901306241,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037379827188544835,
      "loss": 5.1531,
      "mean_token_accuracy": 0.20044768750667571,
      "num_tokens": 82603047.0,
      "step": 39810
    },
    {
      "entropy": 5.558972406387329,
      "epoch": 3.611665457184325,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037376920067481095,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19980100244283677,
      "num_tokens": 82612789.0,
      "step": 39815
    },
    {
      "entropy": 5.485837697982788,
      "epoch": 3.6121190130624092,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003737401274216012,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20356227308511735,
      "num_tokens": 82622881.0,
      "step": 39820
    },
    {
      "entropy": 5.46235466003418,
      "epoch": 3.6125725689404935,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003737110521264206,
      "loss": 5.009,
      "mean_token_accuracy": 0.20379794836044313,
      "num_tokens": 82633933.0,
      "step": 39825
    },
    {
      "entropy": 5.472382545471191,
      "epoch": 3.6130261248185778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003736819747898702,
      "loss": 5.1026,
      "mean_token_accuracy": 0.20366926342248917,
      "num_tokens": 82645069.0,
      "step": 39830
    },
    {
      "entropy": 5.562283134460449,
      "epoch": 3.6134796806966616,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037365289541255154,
      "loss": 5.0924,
      "mean_token_accuracy": 0.20229784548282623,
      "num_tokens": 82655152.0,
      "step": 39835
    },
    {
      "entropy": 5.439989423751831,
      "epoch": 3.613933236574746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037362381399506575,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20424695014953614,
      "num_tokens": 82665260.0,
      "step": 39840
    },
    {
      "entropy": 5.486728096008301,
      "epoch": 3.61438679245283,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037359473053801435,
      "loss": 5.0455,
      "mean_token_accuracy": 0.20666504204273223,
      "num_tokens": 82676287.0,
      "step": 39845
    },
    {
      "entropy": 5.540256834030151,
      "epoch": 3.6148403483309144,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003735656450419988,
      "loss": 5.1687,
      "mean_token_accuracy": 0.19120607376098633,
      "num_tokens": 82686958.0,
      "step": 39850
    },
    {
      "entropy": 5.546939420700073,
      "epoch": 3.6152939042089987,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037353655750762063,
      "loss": 5.221,
      "mean_token_accuracy": 0.1883599951863289,
      "num_tokens": 82698623.0,
      "step": 39855
    },
    {
      "entropy": 5.608243560791015,
      "epoch": 3.615747460087083,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037350746793548135,
      "loss": 5.1337,
      "mean_token_accuracy": 0.20021512508392333,
      "num_tokens": 82709146.0,
      "step": 39860
    },
    {
      "entropy": 5.5096358299255375,
      "epoch": 3.6162010159651667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037347837632618246,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20813469886779784,
      "num_tokens": 82719323.0,
      "step": 39865
    },
    {
      "entropy": 5.491988277435302,
      "epoch": 3.616654571843251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037344928268032574,
      "loss": 5.1145,
      "mean_token_accuracy": 0.2003197655081749,
      "num_tokens": 82730068.0,
      "step": 39870
    },
    {
      "entropy": 5.491295337677002,
      "epoch": 3.6171081277213353,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003734201869985127,
      "loss": 5.1382,
      "mean_token_accuracy": 0.20087573528289795,
      "num_tokens": 82740460.0,
      "step": 39875
    },
    {
      "entropy": 5.52505054473877,
      "epoch": 3.6175616835994195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003733910892813451,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19405651539564134,
      "num_tokens": 82751488.0,
      "step": 39880
    },
    {
      "entropy": 5.445554637908936,
      "epoch": 3.6180152394775034,
      "grad_norm": 0.953125,
      "learning_rate": 0.00037336198952942477,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2140287607908249,
      "num_tokens": 82762061.0,
      "step": 39885
    },
    {
      "entropy": 5.4366577625274655,
      "epoch": 3.6184687953555876,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003733328877433534,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20847563594579696,
      "num_tokens": 82773421.0,
      "step": 39890
    },
    {
      "entropy": 5.353722953796387,
      "epoch": 3.618922351233672,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037330378392373285,
      "loss": 4.8913,
      "mean_token_accuracy": 0.2120847687125206,
      "num_tokens": 82783342.0,
      "step": 39895
    },
    {
      "entropy": 5.466687822341919,
      "epoch": 3.619375907111756,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037327467807116507,
      "loss": 5.0708,
      "mean_token_accuracy": 0.1957324340939522,
      "num_tokens": 82793885.0,
      "step": 39900
    },
    {
      "entropy": 5.5691835403442385,
      "epoch": 3.6198294629898404,
      "grad_norm": 3.359375,
      "learning_rate": 0.00037324557018625186,
      "loss": 5.1183,
      "mean_token_accuracy": 0.20356949269771576,
      "num_tokens": 82803746.0,
      "step": 39905
    },
    {
      "entropy": 5.525136804580688,
      "epoch": 3.6202830188679247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003732164602695952,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20560926645994188,
      "num_tokens": 82813858.0,
      "step": 39910
    },
    {
      "entropy": 5.500539970397949,
      "epoch": 3.620736574746009,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037318734832179705,
      "loss": 5.1103,
      "mean_token_accuracy": 0.2059857130050659,
      "num_tokens": 82824358.0,
      "step": 39915
    },
    {
      "entropy": 5.497749948501587,
      "epoch": 3.621190130624093,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037315823434345963,
      "loss": 5.098,
      "mean_token_accuracy": 0.21150683909654616,
      "num_tokens": 82835801.0,
      "step": 39920
    },
    {
      "entropy": 5.456446409225464,
      "epoch": 3.621643686502177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003731291183351848,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21525522619485854,
      "num_tokens": 82845818.0,
      "step": 39925
    },
    {
      "entropy": 5.460266542434693,
      "epoch": 3.6220972423802613,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037310000029757486,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20313582569360733,
      "num_tokens": 82856920.0,
      "step": 39930
    },
    {
      "entropy": 5.561162233352661,
      "epoch": 3.6225507982583456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003730708802312319,
      "loss": 5.1418,
      "mean_token_accuracy": 0.20255528688430785,
      "num_tokens": 82867220.0,
      "step": 39935
    },
    {
      "entropy": 5.481317043304443,
      "epoch": 3.6230043541364294,
      "grad_norm": 1.0,
      "learning_rate": 0.0003730417581367582,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20941447019577025,
      "num_tokens": 82877541.0,
      "step": 39940
    },
    {
      "entropy": 5.56372480392456,
      "epoch": 3.6234579100145137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003730126340147559,
      "loss": 5.1101,
      "mean_token_accuracy": 0.2081349328160286,
      "num_tokens": 82888346.0,
      "step": 39945
    },
    {
      "entropy": 5.5657110691070555,
      "epoch": 3.623911465892598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003729835078658273,
      "loss": 5.1604,
      "mean_token_accuracy": 0.19876130819320678,
      "num_tokens": 82899721.0,
      "step": 39950
    },
    {
      "entropy": 5.486678647994995,
      "epoch": 3.624365021770682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037295437969057493,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20426099598407746,
      "num_tokens": 82910149.0,
      "step": 39955
    },
    {
      "entropy": 5.539049339294434,
      "epoch": 3.6248185776487665,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037292524948960093,
      "loss": 5.0926,
      "mean_token_accuracy": 0.2016233205795288,
      "num_tokens": 82920883.0,
      "step": 39960
    },
    {
      "entropy": 5.480783987045288,
      "epoch": 3.6252721335268507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003728961172635078,
      "loss": 5.0942,
      "mean_token_accuracy": 0.20244299471378327,
      "num_tokens": 82931091.0,
      "step": 39965
    },
    {
      "entropy": 5.469834852218628,
      "epoch": 3.6257256894049346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037286698301289804,
      "loss": 4.9917,
      "mean_token_accuracy": 0.20449096262454985,
      "num_tokens": 82941108.0,
      "step": 39970
    },
    {
      "entropy": 5.475347661972046,
      "epoch": 3.626179245283019,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0003728378467383742,
      "loss": 5.0652,
      "mean_token_accuracy": 0.2039917841553688,
      "num_tokens": 82951943.0,
      "step": 39975
    },
    {
      "entropy": 5.545006465911865,
      "epoch": 3.626632801161103,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003728087084405387,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19991396814584733,
      "num_tokens": 82961495.0,
      "step": 39980
    },
    {
      "entropy": 5.5750528335571286,
      "epoch": 3.6270863570391874,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037277956811999414,
      "loss": 5.0797,
      "mean_token_accuracy": 0.2053167775273323,
      "num_tokens": 82971858.0,
      "step": 39985
    },
    {
      "entropy": 5.531661081314087,
      "epoch": 3.627539912917271,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003727504257773433,
      "loss": 5.0642,
      "mean_token_accuracy": 0.20202028304338454,
      "num_tokens": 82982805.0,
      "step": 39990
    },
    {
      "entropy": 5.493690061569214,
      "epoch": 3.6279934687953554,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037272128141318866,
      "loss": 5.1,
      "mean_token_accuracy": 0.20017038583755492,
      "num_tokens": 82993446.0,
      "step": 39995
    },
    {
      "entropy": 5.481684350967408,
      "epoch": 3.6284470246734397,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372692135028133,
      "loss": 4.9777,
      "mean_token_accuracy": 0.2051356241106987,
      "num_tokens": 83003378.0,
      "step": 40000
    },
    {
      "entropy": 5.445385885238648,
      "epoch": 3.628900580551524,
      "grad_norm": 1.734375,
      "learning_rate": 0.00037266298662277914,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20214990973472596,
      "num_tokens": 83014223.0,
      "step": 40005
    },
    {
      "entropy": 5.448891115188599,
      "epoch": 3.6293541364296082,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037263383619772986,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19920665472745896,
      "num_tokens": 83024691.0,
      "step": 40010
    },
    {
      "entropy": 5.55369610786438,
      "epoch": 3.6298076923076925,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037260468375358794,
      "loss": 5.0692,
      "mean_token_accuracy": 0.20244454890489577,
      "num_tokens": 83035704.0,
      "step": 40015
    },
    {
      "entropy": 5.577844667434692,
      "epoch": 3.6302612481857763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003725755292909562,
      "loss": 5.1579,
      "mean_token_accuracy": 0.20183795392513276,
      "num_tokens": 83047173.0,
      "step": 40020
    },
    {
      "entropy": 5.532423400878907,
      "epoch": 3.6307148040638606,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003725463728104378,
      "loss": 5.119,
      "mean_token_accuracy": 0.20247270464897155,
      "num_tokens": 83056399.0,
      "step": 40025
    },
    {
      "entropy": 5.429683351516724,
      "epoch": 3.631168359941945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003725172143126354,
      "loss": 5.015,
      "mean_token_accuracy": 0.20213318169116973,
      "num_tokens": 83066413.0,
      "step": 40030
    },
    {
      "entropy": 5.53047661781311,
      "epoch": 3.631621915820029,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037248805379815227,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19966928660869598,
      "num_tokens": 83077843.0,
      "step": 40035
    },
    {
      "entropy": 5.5739185333251955,
      "epoch": 3.632075471698113,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003724588912675913,
      "loss": 5.1611,
      "mean_token_accuracy": 0.20115275233983992,
      "num_tokens": 83088585.0,
      "step": 40040
    },
    {
      "entropy": 5.60074028968811,
      "epoch": 3.6325290275761972,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003724297267215557,
      "loss": 5.2055,
      "mean_token_accuracy": 0.2017166644334793,
      "num_tokens": 83099295.0,
      "step": 40045
    },
    {
      "entropy": 5.493826627731323,
      "epoch": 3.6329825834542815,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037240056016064844,
      "loss": 4.9874,
      "mean_token_accuracy": 0.218414244055748,
      "num_tokens": 83108530.0,
      "step": 40050
    },
    {
      "entropy": 5.4707033157348635,
      "epoch": 3.6334361393323658,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003723713915854728,
      "loss": 5.0659,
      "mean_token_accuracy": 0.20508091449737548,
      "num_tokens": 83119703.0,
      "step": 40055
    },
    {
      "entropy": 5.491539859771729,
      "epoch": 3.63388969521045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003723422209966319,
      "loss": 5.0596,
      "mean_token_accuracy": 0.20490497797727586,
      "num_tokens": 83129323.0,
      "step": 40060
    },
    {
      "entropy": 5.490609169006348,
      "epoch": 3.6343432510885343,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003723130483947292,
      "loss": 5.025,
      "mean_token_accuracy": 0.21117681562900542,
      "num_tokens": 83139339.0,
      "step": 40065
    },
    {
      "entropy": 5.44866795539856,
      "epoch": 3.6347968069666186,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003722838737803678,
      "loss": 5.0321,
      "mean_token_accuracy": 0.2052427500486374,
      "num_tokens": 83149845.0,
      "step": 40070
    },
    {
      "entropy": 5.538512372970581,
      "epoch": 3.6352503628447024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037225469715415106,
      "loss": 5.0818,
      "mean_token_accuracy": 0.1983105033636093,
      "num_tokens": 83160332.0,
      "step": 40075
    },
    {
      "entropy": 5.525948143005371,
      "epoch": 3.6357039187227866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037222551851668245,
      "loss": 5.0975,
      "mean_token_accuracy": 0.2000073030591011,
      "num_tokens": 83170892.0,
      "step": 40080
    },
    {
      "entropy": 5.456927633285522,
      "epoch": 3.636157474600871,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003721963378685652,
      "loss": 5.0309,
      "mean_token_accuracy": 0.2059432089328766,
      "num_tokens": 83181035.0,
      "step": 40085
    },
    {
      "entropy": 5.494335603713989,
      "epoch": 3.6366110304789547,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037216715521040314,
      "loss": 5.0594,
      "mean_token_accuracy": 0.20252211689949035,
      "num_tokens": 83191886.0,
      "step": 40090
    },
    {
      "entropy": 5.492596054077149,
      "epoch": 3.637064586357039,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003721379705427994,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2053829774260521,
      "num_tokens": 83202356.0,
      "step": 40095
    },
    {
      "entropy": 5.471048259735108,
      "epoch": 3.6375181422351233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037210878386635774,
      "loss": 5.1404,
      "mean_token_accuracy": 0.20945202559232712,
      "num_tokens": 83212909.0,
      "step": 40100
    },
    {
      "entropy": 5.539671325683594,
      "epoch": 3.6379716981132075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037207959518168165,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19987807422876358,
      "num_tokens": 83223523.0,
      "step": 40105
    },
    {
      "entropy": 5.592629384994507,
      "epoch": 3.638425253991292,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037205040448937485,
      "loss": 5.1802,
      "mean_token_accuracy": 0.19651449471712112,
      "num_tokens": 83233665.0,
      "step": 40110
    },
    {
      "entropy": 5.420074939727783,
      "epoch": 3.638878809869376,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003720212117900409,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20803346931934358,
      "num_tokens": 83243563.0,
      "step": 40115
    },
    {
      "entropy": 5.475756502151489,
      "epoch": 3.6393323657474603,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037199201708428356,
      "loss": 5.1117,
      "mean_token_accuracy": 0.20201402455568312,
      "num_tokens": 83254246.0,
      "step": 40120
    },
    {
      "entropy": 5.536952972412109,
      "epoch": 3.639785921625544,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003719628203727066,
      "loss": 5.0794,
      "mean_token_accuracy": 0.20048145055770875,
      "num_tokens": 83264361.0,
      "step": 40125
    },
    {
      "entropy": 5.540352249145508,
      "epoch": 3.6402394775036284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003719336216559138,
      "loss": 5.0899,
      "mean_token_accuracy": 0.2047416627407074,
      "num_tokens": 83274193.0,
      "step": 40130
    },
    {
      "entropy": 5.500186491012573,
      "epoch": 3.6406930333817127,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000371904420934509,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20800599306821824,
      "num_tokens": 83284593.0,
      "step": 40135
    },
    {
      "entropy": 5.4344377517700195,
      "epoch": 3.641146589259797,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003718752182090961,
      "loss": 5.0119,
      "mean_token_accuracy": 0.2061515271663666,
      "num_tokens": 83294602.0,
      "step": 40140
    },
    {
      "entropy": 5.463600730895996,
      "epoch": 3.6416001451378808,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000371846013480279,
      "loss": 5.0452,
      "mean_token_accuracy": 0.20725181549787522,
      "num_tokens": 83305399.0,
      "step": 40145
    },
    {
      "entropy": 5.449738836288452,
      "epoch": 3.642053701015965,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037181680674866165,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20194558054208755,
      "num_tokens": 83315696.0,
      "step": 40150
    },
    {
      "entropy": 5.641534757614136,
      "epoch": 3.6425072568940493,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003717875980148481,
      "loss": 5.1975,
      "mean_token_accuracy": 0.19868332147598267,
      "num_tokens": 83326686.0,
      "step": 40155
    },
    {
      "entropy": 5.495759677886963,
      "epoch": 3.6429608127721336,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003717583872794423,
      "loss": 5.0327,
      "mean_token_accuracy": 0.21125292778015137,
      "num_tokens": 83336902.0,
      "step": 40160
    },
    {
      "entropy": 5.443921136856079,
      "epoch": 3.643414368650218,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003717291745430484,
      "loss": 5.0515,
      "mean_token_accuracy": 0.20774907022714614,
      "num_tokens": 83347324.0,
      "step": 40165
    },
    {
      "entropy": 5.41031494140625,
      "epoch": 3.643867924528302,
      "grad_norm": 1.765625,
      "learning_rate": 0.00037169995980627056,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19988386631011962,
      "num_tokens": 83358458.0,
      "step": 40170
    },
    {
      "entropy": 5.463219547271729,
      "epoch": 3.644321480406386,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037167074306971285,
      "loss": 5.0924,
      "mean_token_accuracy": 0.20151455253362655,
      "num_tokens": 83369047.0,
      "step": 40175
    },
    {
      "entropy": 5.504606485366821,
      "epoch": 3.64477503628447,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037164152433397964,
      "loss": 5.0625,
      "mean_token_accuracy": 0.2061273619532585,
      "num_tokens": 83378891.0,
      "step": 40180
    },
    {
      "entropy": 5.508392953872681,
      "epoch": 3.6452285921625545,
      "grad_norm": 1.09375,
      "learning_rate": 0.000371612303599675,
      "loss": 5.1299,
      "mean_token_accuracy": 0.1969906881451607,
      "num_tokens": 83389289.0,
      "step": 40185
    },
    {
      "entropy": 5.585042190551758,
      "epoch": 3.6456821480406387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037158308086740336,
      "loss": 5.1361,
      "mean_token_accuracy": 0.20614794194698333,
      "num_tokens": 83399316.0,
      "step": 40190
    },
    {
      "entropy": 5.491983032226562,
      "epoch": 3.6461357039187225,
      "grad_norm": 1.125,
      "learning_rate": 0.0003715538561377689,
      "loss": 5.0984,
      "mean_token_accuracy": 0.20351441502571105,
      "num_tokens": 83409507.0,
      "step": 40195
    },
    {
      "entropy": 5.55027551651001,
      "epoch": 3.646589259796807,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003715246294113762,
      "loss": 5.1585,
      "mean_token_accuracy": 0.19211129993200302,
      "num_tokens": 83419687.0,
      "step": 40200
    },
    {
      "entropy": 5.622582721710205,
      "epoch": 3.647042815674891,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037149540068882945,
      "loss": 5.1611,
      "mean_token_accuracy": 0.20352017134428024,
      "num_tokens": 83429711.0,
      "step": 40205
    },
    {
      "entropy": 5.505757093429565,
      "epoch": 3.6474963715529753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037146616997073327,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20052348375320433,
      "num_tokens": 83440116.0,
      "step": 40210
    },
    {
      "entropy": 5.476407098770141,
      "epoch": 3.6479499274310596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037143693725769214,
      "loss": 5.0342,
      "mean_token_accuracy": 0.21049815118312837,
      "num_tokens": 83450877.0,
      "step": 40215
    },
    {
      "entropy": 5.40392484664917,
      "epoch": 3.648403483309144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037140770255031066,
      "loss": 4.964,
      "mean_token_accuracy": 0.2097928762435913,
      "num_tokens": 83460080.0,
      "step": 40220
    },
    {
      "entropy": 5.487737512588501,
      "epoch": 3.6488570391872277,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037137846584919316,
      "loss": 5.1065,
      "mean_token_accuracy": 0.20062845200300217,
      "num_tokens": 83469933.0,
      "step": 40225
    },
    {
      "entropy": 5.509954261779785,
      "epoch": 3.649310595065312,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003713492271549445,
      "loss": 5.014,
      "mean_token_accuracy": 0.21154575645923615,
      "num_tokens": 83480477.0,
      "step": 40230
    },
    {
      "entropy": 5.47500057220459,
      "epoch": 3.6497641509433962,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003713199864681693,
      "loss": 5.0614,
      "mean_token_accuracy": 0.20396330654621125,
      "num_tokens": 83491022.0,
      "step": 40235
    },
    {
      "entropy": 5.503693675994873,
      "epoch": 3.6502177068214805,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037129074378947216,
      "loss": 5.1107,
      "mean_token_accuracy": 0.2015614241361618,
      "num_tokens": 83501643.0,
      "step": 40240
    },
    {
      "entropy": 5.568625020980835,
      "epoch": 3.6506712626995643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037126149911945796,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1958999365568161,
      "num_tokens": 83511948.0,
      "step": 40245
    },
    {
      "entropy": 5.510710334777832,
      "epoch": 3.6511248185776486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037123225245873137,
      "loss": 5.0968,
      "mean_token_accuracy": 0.20229997336864472,
      "num_tokens": 83521939.0,
      "step": 40250
    },
    {
      "entropy": 5.575088405609131,
      "epoch": 3.651578374455733,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00037120300380789727,
      "loss": 5.1084,
      "mean_token_accuracy": 0.20100710541009903,
      "num_tokens": 83532098.0,
      "step": 40255
    },
    {
      "entropy": 5.524892377853393,
      "epoch": 3.652031930333817,
      "grad_norm": 0.921875,
      "learning_rate": 0.00037117375316756063,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2050875887274742,
      "num_tokens": 83543238.0,
      "step": 40260
    },
    {
      "entropy": 5.543573379516602,
      "epoch": 3.6524854862119014,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003711445005383263,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19918769896030425,
      "num_tokens": 83552969.0,
      "step": 40265
    },
    {
      "entropy": 5.556288146972657,
      "epoch": 3.6529390420899857,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037111524592079896,
      "loss": 5.1203,
      "mean_token_accuracy": 0.19442883133888245,
      "num_tokens": 83564569.0,
      "step": 40270
    },
    {
      "entropy": 5.540489864349365,
      "epoch": 3.65339259796807,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000371085989315584,
      "loss": 5.1581,
      "mean_token_accuracy": 0.19910785108804702,
      "num_tokens": 83575613.0,
      "step": 40275
    },
    {
      "entropy": 5.545562314987182,
      "epoch": 3.6538461538461537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003710567307232864,
      "loss": 5.1315,
      "mean_token_accuracy": 0.20091571658849716,
      "num_tokens": 83586311.0,
      "step": 40280
    },
    {
      "entropy": 5.557441568374633,
      "epoch": 3.654299709724238,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037102747014451095,
      "loss": 5.1147,
      "mean_token_accuracy": 0.20328834056854247,
      "num_tokens": 83596660.0,
      "step": 40285
    },
    {
      "entropy": 5.534591770172119,
      "epoch": 3.6547532656023223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000370998207579863,
      "loss": 5.1105,
      "mean_token_accuracy": 0.19774965941905975,
      "num_tokens": 83606882.0,
      "step": 40290
    },
    {
      "entropy": 5.588614368438721,
      "epoch": 3.6552068214804065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003709689430299477,
      "loss": 5.1504,
      "mean_token_accuracy": 0.20374949723482133,
      "num_tokens": 83617380.0,
      "step": 40295
    },
    {
      "entropy": 5.546232748031616,
      "epoch": 3.6556603773584904,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003709396764953701,
      "loss": 5.0893,
      "mean_token_accuracy": 0.20344317704439163,
      "num_tokens": 83627961.0,
      "step": 40300
    },
    {
      "entropy": 5.353199481964111,
      "epoch": 3.6561139332365746,
      "grad_norm": 1.0,
      "learning_rate": 0.0003709104079767357,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21590771675109863,
      "num_tokens": 83637646.0,
      "step": 40305
    },
    {
      "entropy": 5.490308618545532,
      "epoch": 3.656567489114659,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037088113747464955,
      "loss": 5.0799,
      "mean_token_accuracy": 0.2082795575261116,
      "num_tokens": 83647807.0,
      "step": 40310
    },
    {
      "entropy": 5.510138273239136,
      "epoch": 3.657021044992743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000370851864989717,
      "loss": 5.104,
      "mean_token_accuracy": 0.19989076554775237,
      "num_tokens": 83658002.0,
      "step": 40315
    },
    {
      "entropy": 5.484579420089721,
      "epoch": 3.6574746008708274,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003708225905225435,
      "loss": 5.036,
      "mean_token_accuracy": 0.2130335420370102,
      "num_tokens": 83667875.0,
      "step": 40320
    },
    {
      "entropy": 5.503822231292725,
      "epoch": 3.6579281567489117,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003707933140737345,
      "loss": 5.0597,
      "mean_token_accuracy": 0.2025639072060585,
      "num_tokens": 83678376.0,
      "step": 40325
    },
    {
      "entropy": 5.519255971908569,
      "epoch": 3.6583817126269955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037076403564389516,
      "loss": 5.1378,
      "mean_token_accuracy": 0.19996340721845626,
      "num_tokens": 83688318.0,
      "step": 40330
    },
    {
      "entropy": 5.493380355834961,
      "epoch": 3.65883526850508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003707347552336313,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19919901937246323,
      "num_tokens": 83698935.0,
      "step": 40335
    },
    {
      "entropy": 5.555589294433593,
      "epoch": 3.659288824383164,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003707054728435482,
      "loss": 5.1712,
      "mean_token_accuracy": 0.2021743103861809,
      "num_tokens": 83709382.0,
      "step": 40340
    },
    {
      "entropy": 5.545590019226074,
      "epoch": 3.6597423802612483,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003706761884742516,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21544511020183563,
      "num_tokens": 83719661.0,
      "step": 40345
    },
    {
      "entropy": 5.56018853187561,
      "epoch": 3.660195936139332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037064690212634707,
      "loss": 5.1453,
      "mean_token_accuracy": 0.19507179409265518,
      "num_tokens": 83730679.0,
      "step": 40350
    },
    {
      "entropy": 5.439592504501343,
      "epoch": 3.6606494920174164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003706176138004401,
      "loss": 4.9819,
      "mean_token_accuracy": 0.2095797061920166,
      "num_tokens": 83740595.0,
      "step": 40355
    },
    {
      "entropy": 5.488222980499268,
      "epoch": 3.6611030478955007,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037058832349713667,
      "loss": 5.1285,
      "mean_token_accuracy": 0.20109139233827591,
      "num_tokens": 83751605.0,
      "step": 40360
    },
    {
      "entropy": 5.572562646865845,
      "epoch": 3.661556603773585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003705590312170422,
      "loss": 5.1777,
      "mean_token_accuracy": 0.19767969399690627,
      "num_tokens": 83763780.0,
      "step": 40365
    },
    {
      "entropy": 5.462290048599243,
      "epoch": 3.662010159651669,
      "grad_norm": 1.0,
      "learning_rate": 0.00037052973696076274,
      "loss": 4.9911,
      "mean_token_accuracy": 0.2099922016263008,
      "num_tokens": 83773800.0,
      "step": 40370
    },
    {
      "entropy": 5.573628997802734,
      "epoch": 3.6624637155297535,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037050044072890395,
      "loss": 5.1286,
      "mean_token_accuracy": 0.18959504067897798,
      "num_tokens": 83784402.0,
      "step": 40375
    },
    {
      "entropy": 5.441457605361938,
      "epoch": 3.6629172714078373,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003704711425220717,
      "loss": 4.9993,
      "mean_token_accuracy": 0.20037677735090256,
      "num_tokens": 83794852.0,
      "step": 40380
    },
    {
      "entropy": 5.520657348632812,
      "epoch": 3.6633708272859216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003704418423408718,
      "loss": 5.0887,
      "mean_token_accuracy": 0.20303571820259095,
      "num_tokens": 83805673.0,
      "step": 40385
    },
    {
      "entropy": 5.49696626663208,
      "epoch": 3.663824383164006,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003704125401859104,
      "loss": 5.0696,
      "mean_token_accuracy": 0.20659293681383134,
      "num_tokens": 83815779.0,
      "step": 40390
    },
    {
      "entropy": 5.564631223678589,
      "epoch": 3.66427793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037038323605779333,
      "loss": 5.0804,
      "mean_token_accuracy": 0.20637603253126144,
      "num_tokens": 83826804.0,
      "step": 40395
    },
    {
      "entropy": 5.503375053405762,
      "epoch": 3.664731494920174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003703539299571266,
      "loss": 5.083,
      "mean_token_accuracy": 0.2015778660774231,
      "num_tokens": 83836500.0,
      "step": 40400
    },
    {
      "entropy": 5.504592561721802,
      "epoch": 3.665185050798258,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003703246218845163,
      "loss": 5.0602,
      "mean_token_accuracy": 0.20437448471784592,
      "num_tokens": 83846647.0,
      "step": 40405
    },
    {
      "entropy": 5.447573041915893,
      "epoch": 3.6656386066763424,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003702953118405686,
      "loss": 5.02,
      "mean_token_accuracy": 0.20930802971124648,
      "num_tokens": 83858088.0,
      "step": 40410
    },
    {
      "entropy": 5.548200607299805,
      "epoch": 3.6660921625544267,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003702659998258895,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20600843727588652,
      "num_tokens": 83868936.0,
      "step": 40415
    },
    {
      "entropy": 5.6104004859924315,
      "epoch": 3.666545718432511,
      "grad_norm": 0.921875,
      "learning_rate": 0.00037023668584108523,
      "loss": 5.199,
      "mean_token_accuracy": 0.1957661420106888,
      "num_tokens": 83879813.0,
      "step": 40420
    },
    {
      "entropy": 5.459425401687622,
      "epoch": 3.6669992743105952,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037020736988676206,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19959584772586822,
      "num_tokens": 83891619.0,
      "step": 40425
    },
    {
      "entropy": 5.532561349868774,
      "epoch": 3.6674528301886795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037017805196352623,
      "loss": 5.1687,
      "mean_token_accuracy": 0.20489844381809236,
      "num_tokens": 83901905.0,
      "step": 40430
    },
    {
      "entropy": 5.468607997894287,
      "epoch": 3.6679063860667633,
      "grad_norm": 1.046875,
      "learning_rate": 0.000370148732071984,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20314668267965316,
      "num_tokens": 83912827.0,
      "step": 40435
    },
    {
      "entropy": 5.447465181350708,
      "epoch": 3.6683599419448476,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003701194102127418,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21272396445274352,
      "num_tokens": 83923719.0,
      "step": 40440
    },
    {
      "entropy": 5.415053272247315,
      "epoch": 3.668813497822932,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003700900863864059,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2122355118393898,
      "num_tokens": 83933715.0,
      "step": 40445
    },
    {
      "entropy": 5.537502765655518,
      "epoch": 3.6692670537010157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003700607605935828,
      "loss": 5.0952,
      "mean_token_accuracy": 0.20286500155925752,
      "num_tokens": 83943946.0,
      "step": 40450
    },
    {
      "entropy": 5.556006383895874,
      "epoch": 3.6697206095791,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003700314328348789,
      "loss": 5.1013,
      "mean_token_accuracy": 0.20471926182508468,
      "num_tokens": 83953414.0,
      "step": 40455
    },
    {
      "entropy": 5.449756622314453,
      "epoch": 3.6701741654571842,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003700021031109008,
      "loss": 5.0034,
      "mean_token_accuracy": 0.2035934165120125,
      "num_tokens": 83963717.0,
      "step": 40460
    },
    {
      "entropy": 5.442269277572632,
      "epoch": 3.6706277213352685,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036997277142225506,
      "loss": 5.057,
      "mean_token_accuracy": 0.206035378575325,
      "num_tokens": 83973467.0,
      "step": 40465
    },
    {
      "entropy": 5.493718767166138,
      "epoch": 3.6710812772133528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036994343776954806,
      "loss": 5.0812,
      "mean_token_accuracy": 0.20808559209108352,
      "num_tokens": 83984013.0,
      "step": 40470
    },
    {
      "entropy": 5.5089105606079105,
      "epoch": 3.671534833091437,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003699141021533867,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20835876017808913,
      "num_tokens": 83994349.0,
      "step": 40475
    },
    {
      "entropy": 5.428650856018066,
      "epoch": 3.6719883889695213,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003698847645743775,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20332825034856797,
      "num_tokens": 84004525.0,
      "step": 40480
    },
    {
      "entropy": 5.4755696773529055,
      "epoch": 3.672441944847605,
      "grad_norm": 1.0,
      "learning_rate": 0.0003698554250331271,
      "loss": 5.0746,
      "mean_token_accuracy": 0.20748146921396254,
      "num_tokens": 84014286.0,
      "step": 40485
    },
    {
      "entropy": 5.4722967624664305,
      "epoch": 3.6728955007256894,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003698260835302424,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20573559701442717,
      "num_tokens": 84024826.0,
      "step": 40490
    },
    {
      "entropy": 5.547454881668091,
      "epoch": 3.6733490566037736,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003697967400663302,
      "loss": 5.0849,
      "mean_token_accuracy": 0.20551309138536453,
      "num_tokens": 84035726.0,
      "step": 40495
    },
    {
      "entropy": 5.445264482498169,
      "epoch": 3.673802612481858,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036976739464199715,
      "loss": 4.9526,
      "mean_token_accuracy": 0.2069300651550293,
      "num_tokens": 84045573.0,
      "step": 40500
    },
    {
      "entropy": 5.547777271270752,
      "epoch": 3.6742561683599417,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036973804725785024,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19736093729734422,
      "num_tokens": 84055983.0,
      "step": 40505
    },
    {
      "entropy": 5.556387853622437,
      "epoch": 3.674709724238026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036970869791449644,
      "loss": 5.1808,
      "mean_token_accuracy": 0.20573480874300004,
      "num_tokens": 84066905.0,
      "step": 40510
    },
    {
      "entropy": 5.433363962173462,
      "epoch": 3.6751632801161103,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036967934661254263,
      "loss": 4.9886,
      "mean_token_accuracy": 0.2149362936615944,
      "num_tokens": 84076861.0,
      "step": 40515
    },
    {
      "entropy": 5.4968830108642575,
      "epoch": 3.6756168359941945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003696499933525957,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20399270653724672,
      "num_tokens": 84086167.0,
      "step": 40520
    },
    {
      "entropy": 5.473440742492675,
      "epoch": 3.676070391872279,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036962063813526287,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20977393984794618,
      "num_tokens": 84095306.0,
      "step": 40525
    },
    {
      "entropy": 5.493261384963989,
      "epoch": 3.676523947750363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003695912809611511,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20395910739898682,
      "num_tokens": 84106505.0,
      "step": 40530
    },
    {
      "entropy": 5.537900924682617,
      "epoch": 3.676977503628447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003695619218308676,
      "loss": 5.035,
      "mean_token_accuracy": 0.205009064078331,
      "num_tokens": 84116393.0,
      "step": 40535
    },
    {
      "entropy": 5.455329942703247,
      "epoch": 3.677431059506531,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036953256074501937,
      "loss": 5.0808,
      "mean_token_accuracy": 0.20264331251382828,
      "num_tokens": 84125890.0,
      "step": 40540
    },
    {
      "entropy": 5.502975177764893,
      "epoch": 3.6778846153846154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003695031977042138,
      "loss": 5.0839,
      "mean_token_accuracy": 0.2068396106362343,
      "num_tokens": 84137736.0,
      "step": 40545
    },
    {
      "entropy": 5.498642683029175,
      "epoch": 3.6783381712626997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036947383270905795,
      "loss": 5.1099,
      "mean_token_accuracy": 0.20338505357503892,
      "num_tokens": 84148974.0,
      "step": 40550
    },
    {
      "entropy": 5.480207061767578,
      "epoch": 3.6787917271407835,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036944446576015914,
      "loss": 5.1084,
      "mean_token_accuracy": 0.2046502336859703,
      "num_tokens": 84159041.0,
      "step": 40555
    },
    {
      "entropy": 5.513012266159057,
      "epoch": 3.6792452830188678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036941509685812476,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21041487008333207,
      "num_tokens": 84169432.0,
      "step": 40560
    },
    {
      "entropy": 5.46841082572937,
      "epoch": 3.679698838896952,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036938572600356213,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20130183845758437,
      "num_tokens": 84179964.0,
      "step": 40565
    },
    {
      "entropy": 5.420202302932739,
      "epoch": 3.6801523947750363,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036935635319707856,
      "loss": 5.0985,
      "mean_token_accuracy": 0.19946988821029663,
      "num_tokens": 84189496.0,
      "step": 40570
    },
    {
      "entropy": 5.502007865905762,
      "epoch": 3.6806059506531206,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003693269784392816,
      "loss": 5.0533,
      "mean_token_accuracy": 0.2148843765258789,
      "num_tokens": 84200396.0,
      "step": 40575
    },
    {
      "entropy": 5.471389722824097,
      "epoch": 3.681059506531205,
      "grad_norm": 1.25,
      "learning_rate": 0.00036929760173077863,
      "loss": 4.9928,
      "mean_token_accuracy": 0.205596262216568,
      "num_tokens": 84210114.0,
      "step": 40580
    },
    {
      "entropy": 5.4689943313598635,
      "epoch": 3.6815130624092887,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003692682230721773,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2040889099240303,
      "num_tokens": 84220472.0,
      "step": 40585
    },
    {
      "entropy": 5.477488613128662,
      "epoch": 3.681966618287373,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003692388424640851,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20581790506839753,
      "num_tokens": 84230718.0,
      "step": 40590
    },
    {
      "entropy": 5.520621204376221,
      "epoch": 3.682420174165457,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036920945990710964,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20728205889463425,
      "num_tokens": 84241736.0,
      "step": 40595
    },
    {
      "entropy": 5.387384128570557,
      "epoch": 3.6828737300435415,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036918007540185846,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20943974256515502,
      "num_tokens": 84251770.0,
      "step": 40600
    },
    {
      "entropy": 5.524249744415283,
      "epoch": 3.6833272859216253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003691506889489394,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20475009232759475,
      "num_tokens": 84263355.0,
      "step": 40605
    },
    {
      "entropy": 5.51814751625061,
      "epoch": 3.6837808417997095,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036912130054896005,
      "loss": 5.1346,
      "mean_token_accuracy": 0.1999964401125908,
      "num_tokens": 84274835.0,
      "step": 40610
    },
    {
      "entropy": 5.564660596847534,
      "epoch": 3.684234397677794,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036909191020252825,
      "loss": 5.178,
      "mean_token_accuracy": 0.189279505610466,
      "num_tokens": 84286099.0,
      "step": 40615
    },
    {
      "entropy": 5.529623985290527,
      "epoch": 3.684687953555878,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003690625179102517,
      "loss": 5.1082,
      "mean_token_accuracy": 0.20749870389699937,
      "num_tokens": 84296168.0,
      "step": 40620
    },
    {
      "entropy": 5.629620695114136,
      "epoch": 3.6851415094339623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003690331236727384,
      "loss": 5.1865,
      "mean_token_accuracy": 0.19426340311765672,
      "num_tokens": 84307352.0,
      "step": 40625
    },
    {
      "entropy": 5.55072512626648,
      "epoch": 3.6855950653120466,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003690037274905961,
      "loss": 5.1107,
      "mean_token_accuracy": 0.20102299898862838,
      "num_tokens": 84318496.0,
      "step": 40630
    },
    {
      "entropy": 5.426001071929932,
      "epoch": 3.686048621190131,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003689743293644327,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20671834498643876,
      "num_tokens": 84329101.0,
      "step": 40635
    },
    {
      "entropy": 5.457547616958618,
      "epoch": 3.6865021770682147,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003689449292948563,
      "loss": 5.0234,
      "mean_token_accuracy": 0.21092824190855025,
      "num_tokens": 84339143.0,
      "step": 40640
    },
    {
      "entropy": 5.516800355911255,
      "epoch": 3.686955732946299,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036891552728247483,
      "loss": 5.0623,
      "mean_token_accuracy": 0.20378196090459824,
      "num_tokens": 84349693.0,
      "step": 40645
    },
    {
      "entropy": 5.489875078201294,
      "epoch": 3.6874092888243832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036888612332789636,
      "loss": 5.0645,
      "mean_token_accuracy": 0.20494346767663957,
      "num_tokens": 84359799.0,
      "step": 40650
    },
    {
      "entropy": 5.488812398910523,
      "epoch": 3.6878628447024675,
      "grad_norm": 1.25,
      "learning_rate": 0.00036885671743172875,
      "loss": 5.0771,
      "mean_token_accuracy": 0.2056769922375679,
      "num_tokens": 84370695.0,
      "step": 40655
    },
    {
      "entropy": 5.501824808120728,
      "epoch": 3.6883164005805513,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003688273095945805,
      "loss": 5.1035,
      "mean_token_accuracy": 0.1990031585097313,
      "num_tokens": 84380205.0,
      "step": 40660
    },
    {
      "entropy": 5.487625455856323,
      "epoch": 3.6887699564586356,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036879789981705954,
      "loss": 5.0599,
      "mean_token_accuracy": 0.20803981870412827,
      "num_tokens": 84391034.0,
      "step": 40665
    },
    {
      "entropy": 5.534577941894531,
      "epoch": 3.68922351233672,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036876848809977404,
      "loss": 5.1476,
      "mean_token_accuracy": 0.1969124436378479,
      "num_tokens": 84401452.0,
      "step": 40670
    },
    {
      "entropy": 5.468833971023559,
      "epoch": 3.689677068214804,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036873907444333234,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20860579758882522,
      "num_tokens": 84411126.0,
      "step": 40675
    },
    {
      "entropy": 5.492573165893555,
      "epoch": 3.6901306240928884,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00036870965884834274,
      "loss": 5.052,
      "mean_token_accuracy": 0.20395248085260392,
      "num_tokens": 84422740.0,
      "step": 40680
    },
    {
      "entropy": 5.507915687561035,
      "epoch": 3.6905841799709727,
      "grad_norm": 1.125,
      "learning_rate": 0.00036868024131541337,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19443824142217636,
      "num_tokens": 84432831.0,
      "step": 40685
    },
    {
      "entropy": 5.516391849517822,
      "epoch": 3.6910377358490565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036865082184515296,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20602402687072754,
      "num_tokens": 84443176.0,
      "step": 40690
    },
    {
      "entropy": 5.558905553817749,
      "epoch": 3.6914912917271407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036862140043816957,
      "loss": 5.1381,
      "mean_token_accuracy": 0.201094551384449,
      "num_tokens": 84453185.0,
      "step": 40695
    },
    {
      "entropy": 5.541849803924561,
      "epoch": 3.691944847605225,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003685919770950718,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20644847750663758,
      "num_tokens": 84464303.0,
      "step": 40700
    },
    {
      "entropy": 5.494391775131225,
      "epoch": 3.6923984034833093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036856255181646797,
      "loss": 5.0909,
      "mean_token_accuracy": 0.2026965320110321,
      "num_tokens": 84474463.0,
      "step": 40705
    },
    {
      "entropy": 5.473930358886719,
      "epoch": 3.692851959361393,
      "grad_norm": 1.0,
      "learning_rate": 0.0003685331246029668,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20682182759046555,
      "num_tokens": 84484682.0,
      "step": 40710
    },
    {
      "entropy": 5.50702862739563,
      "epoch": 3.6933055152394774,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036850369545517685,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20706943720579146,
      "num_tokens": 84494830.0,
      "step": 40715
    },
    {
      "entropy": 5.424637794494629,
      "epoch": 3.6937590711175616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003684742643737065,
      "loss": 5.001,
      "mean_token_accuracy": 0.20704060941934585,
      "num_tokens": 84505449.0,
      "step": 40720
    },
    {
      "entropy": 5.5067853927612305,
      "epoch": 3.694212626995646,
      "grad_norm": 1.0,
      "learning_rate": 0.00036844483135916467,
      "loss": 5.1136,
      "mean_token_accuracy": 0.2032462477684021,
      "num_tokens": 84515833.0,
      "step": 40725
    },
    {
      "entropy": 5.546959686279297,
      "epoch": 3.69466618287373,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003684153964121599,
      "loss": 5.1233,
      "mean_token_accuracy": 0.19750966727733613,
      "num_tokens": 84526677.0,
      "step": 40730
    },
    {
      "entropy": 5.4806946277618405,
      "epoch": 3.6951197387518144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036838595953330095,
      "loss": 5.0434,
      "mean_token_accuracy": 0.21099608540534973,
      "num_tokens": 84536656.0,
      "step": 40735
    },
    {
      "entropy": 5.5156608581542965,
      "epoch": 3.6955732946298983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036835652072319646,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20711729824543,
      "num_tokens": 84546506.0,
      "step": 40740
    },
    {
      "entropy": 5.517757940292358,
      "epoch": 3.6960268505079825,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003683270799824554,
      "loss": 5.1189,
      "mean_token_accuracy": 0.19967640638351442,
      "num_tokens": 84556921.0,
      "step": 40745
    },
    {
      "entropy": 5.497027111053467,
      "epoch": 3.696480406386067,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003682976373116865,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21350231021642685,
      "num_tokens": 84566325.0,
      "step": 40750
    },
    {
      "entropy": 5.391360282897949,
      "epoch": 3.696933962264151,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036826819271149877,
      "loss": 4.9926,
      "mean_token_accuracy": 0.20523724108934402,
      "num_tokens": 84577167.0,
      "step": 40755
    },
    {
      "entropy": 5.418384122848511,
      "epoch": 3.697387518142235,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036823874618250105,
      "loss": 5.024,
      "mean_token_accuracy": 0.20700242966413498,
      "num_tokens": 84586764.0,
      "step": 40760
    },
    {
      "entropy": 5.469256973266601,
      "epoch": 3.697841074020319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036820929772530224,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20610381066799163,
      "num_tokens": 84598903.0,
      "step": 40765
    },
    {
      "entropy": 5.499732589721679,
      "epoch": 3.6982946298984034,
      "grad_norm": 2.03125,
      "learning_rate": 0.0003681798473405114,
      "loss": 5.014,
      "mean_token_accuracy": 0.21243180632591246,
      "num_tokens": 84608695.0,
      "step": 40770
    },
    {
      "entropy": 5.407399892807007,
      "epoch": 3.6987481857764877,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036815039502873756,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21025480031967164,
      "num_tokens": 84619377.0,
      "step": 40775
    },
    {
      "entropy": 5.425534105300903,
      "epoch": 3.699201741654572,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036812094079058985,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20541257858276368,
      "num_tokens": 84629716.0,
      "step": 40780
    },
    {
      "entropy": 5.454553461074829,
      "epoch": 3.699655297532656,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003680914846266773,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21442705541849136,
      "num_tokens": 84640214.0,
      "step": 40785
    },
    {
      "entropy": 5.514241313934326,
      "epoch": 3.7001088534107405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003680620265376092,
      "loss": 5.124,
      "mean_token_accuracy": 0.20345288664102554,
      "num_tokens": 84650403.0,
      "step": 40790
    },
    {
      "entropy": 5.518323087692261,
      "epoch": 3.7005624092888243,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036803256652399456,
      "loss": 5.0093,
      "mean_token_accuracy": 0.203845077753067,
      "num_tokens": 84660912.0,
      "step": 40795
    },
    {
      "entropy": 5.484733200073242,
      "epoch": 3.7010159651669086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036800310458644285,
      "loss": 5.0841,
      "mean_token_accuracy": 0.19756557792425156,
      "num_tokens": 84672194.0,
      "step": 40800
    },
    {
      "entropy": 5.503572225570679,
      "epoch": 3.701469521044993,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036797364072556316,
      "loss": 5.1439,
      "mean_token_accuracy": 0.19829245507717133,
      "num_tokens": 84681850.0,
      "step": 40805
    },
    {
      "entropy": 5.519626140594482,
      "epoch": 3.7019230769230766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000367944174941965,
      "loss": 5.0546,
      "mean_token_accuracy": 0.21135856062173844,
      "num_tokens": 84691271.0,
      "step": 40810
    },
    {
      "entropy": 5.463578987121582,
      "epoch": 3.702376632801161,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036791470723625746,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20098462402820588,
      "num_tokens": 84702189.0,
      "step": 40815
    },
    {
      "entropy": 5.45005989074707,
      "epoch": 3.702830188679245,
      "grad_norm": 1.125,
      "learning_rate": 0.0003678852376090502,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2198386609554291,
      "num_tokens": 84712399.0,
      "step": 40820
    },
    {
      "entropy": 5.522965383529663,
      "epoch": 3.7032837445573294,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003678557660609525,
      "loss": 5.1695,
      "mean_token_accuracy": 0.20029887706041336,
      "num_tokens": 84723174.0,
      "step": 40825
    },
    {
      "entropy": 5.484543371200561,
      "epoch": 3.7037373004354137,
      "grad_norm": 1.0,
      "learning_rate": 0.000367826292592574,
      "loss": 5.0424,
      "mean_token_accuracy": 0.21370830833911897,
      "num_tokens": 84732498.0,
      "step": 40830
    },
    {
      "entropy": 5.514673709869385,
      "epoch": 3.704190856313498,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003677968172045241,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20357510596513748,
      "num_tokens": 84742362.0,
      "step": 40835
    },
    {
      "entropy": 5.481372213363647,
      "epoch": 3.7046444121915822,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003677673398974122,
      "loss": 5.059,
      "mean_token_accuracy": 0.2052364692091942,
      "num_tokens": 84753379.0,
      "step": 40840
    },
    {
      "entropy": 5.554949140548706,
      "epoch": 3.705097968069666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003677378606718482,
      "loss": 5.1692,
      "mean_token_accuracy": 0.1982421800494194,
      "num_tokens": 84763245.0,
      "step": 40845
    },
    {
      "entropy": 5.532168769836426,
      "epoch": 3.7055515239477503,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0003677083795284416,
      "loss": 5.0951,
      "mean_token_accuracy": 0.19988025575876237,
      "num_tokens": 84773929.0,
      "step": 40850
    },
    {
      "entropy": 5.547180938720703,
      "epoch": 3.7060050798258346,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036767889646780204,
      "loss": 5.0985,
      "mean_token_accuracy": 0.19641469269990922,
      "num_tokens": 84784733.0,
      "step": 40855
    },
    {
      "entropy": 5.466436815261841,
      "epoch": 3.706458635703919,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003676494114905394,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20339967608451842,
      "num_tokens": 84794254.0,
      "step": 40860
    },
    {
      "entropy": 5.50618543624878,
      "epoch": 3.7069121915820027,
      "grad_norm": 1.25,
      "learning_rate": 0.00036761992459726325,
      "loss": 5.075,
      "mean_token_accuracy": 0.19783280789852142,
      "num_tokens": 84804361.0,
      "step": 40865
    },
    {
      "entropy": 5.507550811767578,
      "epoch": 3.707365747460087,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036759043578858344,
      "loss": 5.0363,
      "mean_token_accuracy": 0.21407384276390076,
      "num_tokens": 84814497.0,
      "step": 40870
    },
    {
      "entropy": 5.571894550323487,
      "epoch": 3.7078193033381712,
      "grad_norm": 2.203125,
      "learning_rate": 0.00036756094506510984,
      "loss": 5.0715,
      "mean_token_accuracy": 0.20023233443498611,
      "num_tokens": 84826871.0,
      "step": 40875
    },
    {
      "entropy": 5.518542242050171,
      "epoch": 3.7082728592162555,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036753145242745226,
      "loss": 5.0688,
      "mean_token_accuracy": 0.20104046612977983,
      "num_tokens": 84837053.0,
      "step": 40880
    },
    {
      "entropy": 5.432530927658081,
      "epoch": 3.7087264150943398,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003675019578762208,
      "loss": 4.9966,
      "mean_token_accuracy": 0.21159861832857133,
      "num_tokens": 84846681.0,
      "step": 40885
    },
    {
      "entropy": 5.435589456558228,
      "epoch": 3.709179970972424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003674724614120251,
      "loss": 4.9772,
      "mean_token_accuracy": 0.210668708384037,
      "num_tokens": 84856651.0,
      "step": 40890
    },
    {
      "entropy": 5.503957223892212,
      "epoch": 3.709633526850508,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036744296303547547,
      "loss": 5.1525,
      "mean_token_accuracy": 0.1973390743136406,
      "num_tokens": 84867531.0,
      "step": 40895
    },
    {
      "entropy": 5.468797731399536,
      "epoch": 3.710087082728592,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003674134627471818,
      "loss": 5.0315,
      "mean_token_accuracy": 0.21078903079032899,
      "num_tokens": 84879132.0,
      "step": 40900
    },
    {
      "entropy": 5.497129917144775,
      "epoch": 3.7105406386066764,
      "grad_norm": 1.0625,
      "learning_rate": 0.000367383960547754,
      "loss": 5.0751,
      "mean_token_accuracy": 0.21310591250658034,
      "num_tokens": 84888873.0,
      "step": 40905
    },
    {
      "entropy": 5.485200262069702,
      "epoch": 3.7109941944847606,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036735445643780256,
      "loss": 5.0972,
      "mean_token_accuracy": 0.20194421112537383,
      "num_tokens": 84898926.0,
      "step": 40910
    },
    {
      "entropy": 5.559149360656738,
      "epoch": 3.7114477503628445,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003673249504179374,
      "loss": 5.0604,
      "mean_token_accuracy": 0.20588236004114152,
      "num_tokens": 84908502.0,
      "step": 40915
    },
    {
      "entropy": 5.513593864440918,
      "epoch": 3.7119013062409287,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003672954424887687,
      "loss": 5.0673,
      "mean_token_accuracy": 0.20426676273345948,
      "num_tokens": 84919681.0,
      "step": 40920
    },
    {
      "entropy": 5.460544204711914,
      "epoch": 3.712354862119013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003672659326509066,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20832482874393463,
      "num_tokens": 84930149.0,
      "step": 40925
    },
    {
      "entropy": 5.442283868789673,
      "epoch": 3.7128084179970973,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036723642090496166,
      "loss": 5.0062,
      "mean_token_accuracy": 0.21131190806627273,
      "num_tokens": 84940086.0,
      "step": 40930
    },
    {
      "entropy": 5.512076663970947,
      "epoch": 3.7132619738751815,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036720690725154404,
      "loss": 5.0914,
      "mean_token_accuracy": 0.20698213428258896,
      "num_tokens": 84949248.0,
      "step": 40935
    },
    {
      "entropy": 5.462844562530518,
      "epoch": 3.713715529753266,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036717739169126395,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20902120620012282,
      "num_tokens": 84960892.0,
      "step": 40940
    },
    {
      "entropy": 5.4850671768188475,
      "epoch": 3.7141690856313496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036714787422473206,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20917603820562364,
      "num_tokens": 84971024.0,
      "step": 40945
    },
    {
      "entropy": 5.360691738128662,
      "epoch": 3.714622641509434,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003671183548525586,
      "loss": 4.9638,
      "mean_token_accuracy": 0.2095506951212883,
      "num_tokens": 84980945.0,
      "step": 40950
    },
    {
      "entropy": 5.495189619064331,
      "epoch": 3.715076197387518,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000367088833575354,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20057665854692458,
      "num_tokens": 84992055.0,
      "step": 40955
    },
    {
      "entropy": 5.540802955627441,
      "epoch": 3.7155297532656024,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036705931039372904,
      "loss": 5.0755,
      "mean_token_accuracy": 0.2069353550672531,
      "num_tokens": 85002619.0,
      "step": 40960
    },
    {
      "entropy": 5.496883726119995,
      "epoch": 3.7159833091436862,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003670297853082939,
      "loss": 5.058,
      "mean_token_accuracy": 0.20454563945531845,
      "num_tokens": 85013317.0,
      "step": 40965
    },
    {
      "entropy": 5.464252710342407,
      "epoch": 3.7164368650217705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036700025831965937,
      "loss": 5.0321,
      "mean_token_accuracy": 0.2041796013712883,
      "num_tokens": 85023457.0,
      "step": 40970
    },
    {
      "entropy": 5.500439023971557,
      "epoch": 3.7168904208998548,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036697072942843616,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20138485729694366,
      "num_tokens": 85033498.0,
      "step": 40975
    },
    {
      "entropy": 5.668953561782837,
      "epoch": 3.717343976777939,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003669411986352348,
      "loss": 5.2663,
      "mean_token_accuracy": 0.19858208000659944,
      "num_tokens": 85043687.0,
      "step": 40980
    },
    {
      "entropy": 5.552313899993896,
      "epoch": 3.7177975326560233,
      "grad_norm": 1.1875,
      "learning_rate": 0.000366911665940666,
      "loss": 5.1237,
      "mean_token_accuracy": 0.2019405946135521,
      "num_tokens": 85055045.0,
      "step": 40985
    },
    {
      "entropy": 5.45524001121521,
      "epoch": 3.7182510885341076,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036688213134534044,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21450158059597016,
      "num_tokens": 85064625.0,
      "step": 40990
    },
    {
      "entropy": 5.577834224700927,
      "epoch": 3.718704644412192,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003668525948498691,
      "loss": 5.2245,
      "mean_token_accuracy": 0.18914433419704438,
      "num_tokens": 85075249.0,
      "step": 40995
    },
    {
      "entropy": 5.5143403053283695,
      "epoch": 3.7191582002902757,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036682305645486266,
      "loss": 5.0702,
      "mean_token_accuracy": 0.20590040534734727,
      "num_tokens": 85085900.0,
      "step": 41000
    },
    {
      "entropy": 5.506842947006225,
      "epoch": 3.71961175616836,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036679351616093194,
      "loss": 5.083,
      "mean_token_accuracy": 0.20606561303138732,
      "num_tokens": 85095772.0,
      "step": 41005
    },
    {
      "entropy": 5.49042911529541,
      "epoch": 3.720065312046444,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036676397396868797,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20841867327690125,
      "num_tokens": 85105653.0,
      "step": 41010
    },
    {
      "entropy": 5.524948167800903,
      "epoch": 3.7205188679245285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003667344298787417,
      "loss": 5.1127,
      "mean_token_accuracy": 0.20259731262922287,
      "num_tokens": 85115432.0,
      "step": 41015
    },
    {
      "entropy": 5.566104888916016,
      "epoch": 3.7209724238026123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036670488389170395,
      "loss": 5.2646,
      "mean_token_accuracy": 0.19085694700479508,
      "num_tokens": 85126572.0,
      "step": 41020
    },
    {
      "entropy": 5.465870952606201,
      "epoch": 3.7214259796806965,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036667533600818586,
      "loss": 4.9961,
      "mean_token_accuracy": 0.21108545809984208,
      "num_tokens": 85136809.0,
      "step": 41025
    },
    {
      "entropy": 5.71265139579773,
      "epoch": 3.721879535558781,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003666457862287985,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18830746412277222,
      "num_tokens": 85148346.0,
      "step": 41030
    },
    {
      "entropy": 5.50457239151001,
      "epoch": 3.722333091436865,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00036661623455415296,
      "loss": 5.0529,
      "mean_token_accuracy": 0.2061544805765152,
      "num_tokens": 85158807.0,
      "step": 41035
    },
    {
      "entropy": 5.528996181488037,
      "epoch": 3.7227866473149493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003665866809848602,
      "loss": 5.1755,
      "mean_token_accuracy": 0.19449986517429352,
      "num_tokens": 85170752.0,
      "step": 41040
    },
    {
      "entropy": 5.4588476657867435,
      "epoch": 3.7232402031930336,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036655712552153164,
      "loss": 5.072,
      "mean_token_accuracy": 0.2024569883942604,
      "num_tokens": 85180779.0,
      "step": 41045
    },
    {
      "entropy": 5.532368803024292,
      "epoch": 3.7236937590711174,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003665275681647784,
      "loss": 5.0626,
      "mean_token_accuracy": 0.20965212136507033,
      "num_tokens": 85191391.0,
      "step": 41050
    },
    {
      "entropy": 5.526321649551392,
      "epoch": 3.7241473149492017,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003664980089152116,
      "loss": 5.0275,
      "mean_token_accuracy": 0.2039183795452118,
      "num_tokens": 85201357.0,
      "step": 41055
    },
    {
      "entropy": 5.444402837753296,
      "epoch": 3.724600870827286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036646844777344284,
      "loss": 5.072,
      "mean_token_accuracy": 0.20132942199707032,
      "num_tokens": 85212193.0,
      "step": 41060
    },
    {
      "entropy": 5.5361230850219725,
      "epoch": 3.7250544267053702,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036643888474008313,
      "loss": 5.0875,
      "mean_token_accuracy": 0.2049831137061119,
      "num_tokens": 85223809.0,
      "step": 41065
    },
    {
      "entropy": 5.537462329864502,
      "epoch": 3.725507982583454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003664093198157441,
      "loss": 5.0401,
      "mean_token_accuracy": 0.2085432916879654,
      "num_tokens": 85234992.0,
      "step": 41070
    },
    {
      "entropy": 5.538163566589356,
      "epoch": 3.7259615384615383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003663797530010369,
      "loss": 5.1029,
      "mean_token_accuracy": 0.20200563669204713,
      "num_tokens": 85244157.0,
      "step": 41075
    },
    {
      "entropy": 5.460529136657715,
      "epoch": 3.7264150943396226,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003663501842965732,
      "loss": 5.0222,
      "mean_token_accuracy": 0.21009781062602997,
      "num_tokens": 85255946.0,
      "step": 41080
    },
    {
      "entropy": 5.438874292373657,
      "epoch": 3.726868650217707,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003663206137029645,
      "loss": 5.0123,
      "mean_token_accuracy": 0.21225429326295853,
      "num_tokens": 85265273.0,
      "step": 41085
    },
    {
      "entropy": 5.391030979156494,
      "epoch": 3.727322206095791,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036629104122082215,
      "loss": 4.8718,
      "mean_token_accuracy": 0.215790319442749,
      "num_tokens": 85275794.0,
      "step": 41090
    },
    {
      "entropy": 5.491257238388061,
      "epoch": 3.7277757619738754,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036626146685075775,
      "loss": 5.1701,
      "mean_token_accuracy": 0.19388416558504104,
      "num_tokens": 85287334.0,
      "step": 41095
    },
    {
      "entropy": 5.5304200649261475,
      "epoch": 3.728229317851959,
      "grad_norm": 1.0625,
      "learning_rate": 0.000366231890593383,
      "loss": 5.0628,
      "mean_token_accuracy": 0.20835942924022674,
      "num_tokens": 85297320.0,
      "step": 41100
    },
    {
      "entropy": 5.444436931610108,
      "epoch": 3.7286828737300435,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003662023124493095,
      "loss": 4.9814,
      "mean_token_accuracy": 0.21137585490942,
      "num_tokens": 85308071.0,
      "step": 41105
    },
    {
      "entropy": 5.498539590835572,
      "epoch": 3.7291364296081277,
      "grad_norm": 0.984375,
      "learning_rate": 0.000366172732419149,
      "loss": 5.0554,
      "mean_token_accuracy": 0.20230919867753983,
      "num_tokens": 85317790.0,
      "step": 41110
    },
    {
      "entropy": 5.530544281005859,
      "epoch": 3.729589985486212,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036614315050351317,
      "loss": 5.1454,
      "mean_token_accuracy": 0.19936541467905045,
      "num_tokens": 85329008.0,
      "step": 41115
    },
    {
      "entropy": 5.607451152801514,
      "epoch": 3.730043541364296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003661135667030137,
      "loss": 5.183,
      "mean_token_accuracy": 0.19518521428108215,
      "num_tokens": 85339095.0,
      "step": 41120
    },
    {
      "entropy": 5.592536354064942,
      "epoch": 3.73049709724238,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003660839810182624,
      "loss": 5.1266,
      "mean_token_accuracy": 0.19921765029430388,
      "num_tokens": 85349833.0,
      "step": 41125
    },
    {
      "entropy": 5.519586563110352,
      "epoch": 3.7309506531204644,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00036605439344987125,
      "loss": 5.1185,
      "mean_token_accuracy": 0.20297229290008545,
      "num_tokens": 85359512.0,
      "step": 41130
    },
    {
      "entropy": 5.469330167770385,
      "epoch": 3.7314042089985486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000366024803998452,
      "loss": 5.0992,
      "mean_token_accuracy": 0.20954614132642746,
      "num_tokens": 85370053.0,
      "step": 41135
    },
    {
      "entropy": 5.48724889755249,
      "epoch": 3.731857764876633,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036599521266461653,
      "loss": 5.0554,
      "mean_token_accuracy": 0.20590486973524094,
      "num_tokens": 85380020.0,
      "step": 41140
    },
    {
      "entropy": 5.520630311965943,
      "epoch": 3.732311320754717,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036596561944897703,
      "loss": 5.0626,
      "mean_token_accuracy": 0.20328411161899568,
      "num_tokens": 85390232.0,
      "step": 41145
    },
    {
      "entropy": 5.397369337081909,
      "epoch": 3.7327648766328014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003659360243521453,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21566662937402725,
      "num_tokens": 85400401.0,
      "step": 41150
    },
    {
      "entropy": 5.487556600570679,
      "epoch": 3.7332184325108853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003659064273747334,
      "loss": 5.0964,
      "mean_token_accuracy": 0.20453782081604005,
      "num_tokens": 85411749.0,
      "step": 41155
    },
    {
      "entropy": 5.619144010543823,
      "epoch": 3.7336719883889695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036587682851735336,
      "loss": 5.1749,
      "mean_token_accuracy": 0.1960899665951729,
      "num_tokens": 85424065.0,
      "step": 41160
    },
    {
      "entropy": 5.522029638290405,
      "epoch": 3.734125544267054,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036584722778061743,
      "loss": 5.0874,
      "mean_token_accuracy": 0.20500210374593736,
      "num_tokens": 85434412.0,
      "step": 41165
    },
    {
      "entropy": 5.512126922607422,
      "epoch": 3.7345791001451376,
      "grad_norm": 1.0,
      "learning_rate": 0.0003658176251651376,
      "loss": 5.059,
      "mean_token_accuracy": 0.21010550260543823,
      "num_tokens": 85444489.0,
      "step": 41170
    },
    {
      "entropy": 5.518207168579101,
      "epoch": 3.735032656023222,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036578802067152615,
      "loss": 5.0542,
      "mean_token_accuracy": 0.2050297662615776,
      "num_tokens": 85454556.0,
      "step": 41175
    },
    {
      "entropy": 5.4244519710540775,
      "epoch": 3.735486211901306,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003657584143003954,
      "loss": 4.9842,
      "mean_token_accuracy": 0.2109931603074074,
      "num_tokens": 85464701.0,
      "step": 41180
    },
    {
      "entropy": 5.492675685882569,
      "epoch": 3.7359397677793904,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003657288060523574,
      "loss": 5.1248,
      "mean_token_accuracy": 0.20696713775396347,
      "num_tokens": 85475657.0,
      "step": 41185
    },
    {
      "entropy": 5.523292970657349,
      "epoch": 3.7363933236574747,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036569919592802466,
      "loss": 5.0602,
      "mean_token_accuracy": 0.21085490882396699,
      "num_tokens": 85486358.0,
      "step": 41190
    },
    {
      "entropy": 5.574515247344971,
      "epoch": 3.736846879535559,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003656695839280093,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19472882449626921,
      "num_tokens": 85497913.0,
      "step": 41195
    },
    {
      "entropy": 5.516621398925781,
      "epoch": 3.737300435413643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036563997005292396,
      "loss": 5.0709,
      "mean_token_accuracy": 0.2112588405609131,
      "num_tokens": 85508314.0,
      "step": 41200
    },
    {
      "entropy": 5.455701732635498,
      "epoch": 3.737753991291727,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036561035430338084,
      "loss": 5.0544,
      "mean_token_accuracy": 0.20390921384096145,
      "num_tokens": 85518142.0,
      "step": 41205
    },
    {
      "entropy": 5.447376441955567,
      "epoch": 3.7382075471698113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036558073667999257,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21249291449785232,
      "num_tokens": 85526854.0,
      "step": 41210
    },
    {
      "entropy": 5.3950817584991455,
      "epoch": 3.7386611030478956,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036555111718337166,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21751340627670288,
      "num_tokens": 85537153.0,
      "step": 41215
    },
    {
      "entropy": 5.552774810791016,
      "epoch": 3.73911465892598,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003655214958141305,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19181670397520065,
      "num_tokens": 85548583.0,
      "step": 41220
    },
    {
      "entropy": 5.531194448471069,
      "epoch": 3.7395682148040637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036549187257288164,
      "loss": 5.0944,
      "mean_token_accuracy": 0.21359411925077437,
      "num_tokens": 85559228.0,
      "step": 41225
    },
    {
      "entropy": 5.521360111236572,
      "epoch": 3.740021770682148,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036546224746023795,
      "loss": 5.0535,
      "mean_token_accuracy": 0.20436348021030426,
      "num_tokens": 85569757.0,
      "step": 41230
    },
    {
      "entropy": 5.476427745819092,
      "epoch": 3.740475326560232,
      "grad_norm": 1.0,
      "learning_rate": 0.00036543262047681186,
      "loss": 5.0929,
      "mean_token_accuracy": 0.20959011912345887,
      "num_tokens": 85582013.0,
      "step": 41235
    },
    {
      "entropy": 5.477881336212159,
      "epoch": 3.7409288824383164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003654029916232161,
      "loss": 5.064,
      "mean_token_accuracy": 0.2192763105034828,
      "num_tokens": 85594049.0,
      "step": 41240
    },
    {
      "entropy": 5.53540186882019,
      "epoch": 3.7413824383164007,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036537336090006347,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19908110946416854,
      "num_tokens": 85605253.0,
      "step": 41245
    },
    {
      "entropy": 5.486790800094605,
      "epoch": 3.741835994194485,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036534372830796674,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21172345131635667,
      "num_tokens": 85615032.0,
      "step": 41250
    },
    {
      "entropy": 5.438778305053711,
      "epoch": 3.742289550072569,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003653140938475386,
      "loss": 5.0456,
      "mean_token_accuracy": 0.21420425176620483,
      "num_tokens": 85625497.0,
      "step": 41255
    },
    {
      "entropy": 5.482673692703247,
      "epoch": 3.742743105950653,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00036528445751939206,
      "loss": 5.084,
      "mean_token_accuracy": 0.2051186352968216,
      "num_tokens": 85636058.0,
      "step": 41260
    },
    {
      "entropy": 5.520774173736572,
      "epoch": 3.7431966618287373,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036525481932413997,
      "loss": 5.0885,
      "mean_token_accuracy": 0.20124469995498656,
      "num_tokens": 85646505.0,
      "step": 41265
    },
    {
      "entropy": 5.569367170333862,
      "epoch": 3.7436502177068216,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036522517926239513,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19963208734989166,
      "num_tokens": 85656864.0,
      "step": 41270
    },
    {
      "entropy": 5.4411482334136965,
      "epoch": 3.7441037735849054,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00036519553733477055,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21738671511411667,
      "num_tokens": 85666783.0,
      "step": 41275
    },
    {
      "entropy": 5.37396502494812,
      "epoch": 3.7445573294629897,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003651658935418794,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21677025109529496,
      "num_tokens": 85677102.0,
      "step": 41280
    },
    {
      "entropy": 5.445438432693481,
      "epoch": 3.745010885341074,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036513624788433454,
      "loss": 5.003,
      "mean_token_accuracy": 0.21468648612499236,
      "num_tokens": 85687048.0,
      "step": 41285
    },
    {
      "entropy": 5.511070299148559,
      "epoch": 3.7454644412191582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003651066003627491,
      "loss": 5.0798,
      "mean_token_accuracy": 0.1953260764479637,
      "num_tokens": 85698314.0,
      "step": 41290
    },
    {
      "entropy": 5.540169382095337,
      "epoch": 3.7459179970972425,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003650769509777362,
      "loss": 5.1314,
      "mean_token_accuracy": 0.19825348109006882,
      "num_tokens": 85708039.0,
      "step": 41295
    },
    {
      "entropy": 5.472524118423462,
      "epoch": 3.7463715529753268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036504729972990904,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21285380721092223,
      "num_tokens": 85717741.0,
      "step": 41300
    },
    {
      "entropy": 5.526195955276489,
      "epoch": 3.7468251088534106,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003650176466198807,
      "loss": 5.0432,
      "mean_token_accuracy": 0.20882771909236908,
      "num_tokens": 85728286.0,
      "step": 41305
    },
    {
      "entropy": 5.489229917526245,
      "epoch": 3.747278664731495,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036498799164826453,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20937821865081788,
      "num_tokens": 85738970.0,
      "step": 41310
    },
    {
      "entropy": 5.417154121398926,
      "epoch": 3.747732220609579,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036495833481567375,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21134013086557388,
      "num_tokens": 85749284.0,
      "step": 41315
    },
    {
      "entropy": 5.5254124164581295,
      "epoch": 3.7481857764876634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036492867612272176,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19778684228658677,
      "num_tokens": 85759595.0,
      "step": 41320
    },
    {
      "entropy": 5.588000869750976,
      "epoch": 3.748639332365747,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003648990155700217,
      "loss": 5.1665,
      "mean_token_accuracy": 0.19665064066648483,
      "num_tokens": 85771049.0,
      "step": 41325
    },
    {
      "entropy": 5.481721687316894,
      "epoch": 3.7490928882438315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036486935315818713,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20028374046087266,
      "num_tokens": 85781656.0,
      "step": 41330
    },
    {
      "entropy": 5.398364400863647,
      "epoch": 3.7495464441219157,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036483968888783154,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21132886409759521,
      "num_tokens": 85790922.0,
      "step": 41335
    },
    {
      "entropy": 5.460790061950684,
      "epoch": 3.75,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036481002275956813,
      "loss": 5.003,
      "mean_token_accuracy": 0.21774756610393525,
      "num_tokens": 85799928.0,
      "step": 41340
    },
    {
      "entropy": 5.457953882217407,
      "epoch": 3.7504535558780843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003647803547740106,
      "loss": 5.0782,
      "mean_token_accuracy": 0.2002843514084816,
      "num_tokens": 85810594.0,
      "step": 41345
    },
    {
      "entropy": 5.474414539337158,
      "epoch": 3.7509071117561685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036475068493177254,
      "loss": 5.0591,
      "mean_token_accuracy": 0.20038425475358962,
      "num_tokens": 85820690.0,
      "step": 41350
    },
    {
      "entropy": 5.471064281463623,
      "epoch": 3.751360667634253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003647210132334674,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20875232219696044,
      "num_tokens": 85830314.0,
      "step": 41355
    },
    {
      "entropy": 5.489317798614502,
      "epoch": 3.7518142235123366,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003646913396797088,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19614251255989074,
      "num_tokens": 85841099.0,
      "step": 41360
    },
    {
      "entropy": 5.543775033950806,
      "epoch": 3.752267779390421,
      "grad_norm": 0.890625,
      "learning_rate": 0.00036466166427111046,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20576069205999375,
      "num_tokens": 85852813.0,
      "step": 41365
    },
    {
      "entropy": 5.470551919937134,
      "epoch": 3.752721335268505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036463198700828616,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2093653306365013,
      "num_tokens": 85862778.0,
      "step": 41370
    },
    {
      "entropy": 5.4404158115386965,
      "epoch": 3.7531748911465894,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003646023078918494,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20911163836717606,
      "num_tokens": 85872729.0,
      "step": 41375
    },
    {
      "entropy": 5.517641496658325,
      "epoch": 3.7536284470246732,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003645726269224141,
      "loss": 5.0562,
      "mean_token_accuracy": 0.21063999384641646,
      "num_tokens": 85883289.0,
      "step": 41380
    },
    {
      "entropy": 5.443127870559692,
      "epoch": 3.7540820029027575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036454294410059407,
      "loss": 4.9552,
      "mean_token_accuracy": 0.22213456332683562,
      "num_tokens": 85892318.0,
      "step": 41385
    },
    {
      "entropy": 5.406980037689209,
      "epoch": 3.7545355587808418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003645132594270032,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20622369647026062,
      "num_tokens": 85902719.0,
      "step": 41390
    },
    {
      "entropy": 5.446575164794922,
      "epoch": 3.754989114658926,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036448357290225523,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20645055472850798,
      "num_tokens": 85913271.0,
      "step": 41395
    },
    {
      "entropy": 5.4462768077850345,
      "epoch": 3.7554426705370103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003644538845269642,
      "loss": 4.95,
      "mean_token_accuracy": 0.2139722838997841,
      "num_tokens": 85922397.0,
      "step": 41400
    },
    {
      "entropy": 5.503204679489135,
      "epoch": 3.7558962264150946,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000364424194301744,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20362812280654907,
      "num_tokens": 85933261.0,
      "step": 41405
    },
    {
      "entropy": 5.531245183944702,
      "epoch": 3.7563497822931784,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003643945022272087,
      "loss": 5.1413,
      "mean_token_accuracy": 0.20626943409442902,
      "num_tokens": 85943860.0,
      "step": 41410
    },
    {
      "entropy": 5.540589714050293,
      "epoch": 3.7568033381712627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003643648083039724,
      "loss": 5.1173,
      "mean_token_accuracy": 0.19978623241186141,
      "num_tokens": 85953984.0,
      "step": 41415
    },
    {
      "entropy": 5.495627450942993,
      "epoch": 3.757256894049347,
      "grad_norm": 1.046875,
      "learning_rate": 0.000364335112532649,
      "loss": 5.0488,
      "mean_token_accuracy": 0.21152379363775253,
      "num_tokens": 85964771.0,
      "step": 41420
    },
    {
      "entropy": 5.491361808776856,
      "epoch": 3.757710449927431,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003643054149138527,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19989572763442992,
      "num_tokens": 85975859.0,
      "step": 41425
    },
    {
      "entropy": 5.60704402923584,
      "epoch": 3.758164005805515,
      "grad_norm": 1.0,
      "learning_rate": 0.00036427571544819765,
      "loss": 5.1464,
      "mean_token_accuracy": 0.20078708827495576,
      "num_tokens": 85987022.0,
      "step": 41430
    },
    {
      "entropy": 5.468153619766236,
      "epoch": 3.7586175616835993,
      "grad_norm": 1.0,
      "learning_rate": 0.0003642460141362981,
      "loss": 4.9923,
      "mean_token_accuracy": 0.21076902151107788,
      "num_tokens": 85997056.0,
      "step": 41435
    },
    {
      "entropy": 5.504251146316529,
      "epoch": 3.7590711175616836,
      "grad_norm": 1.0,
      "learning_rate": 0.00036421631097876823,
      "loss": 5.1092,
      "mean_token_accuracy": 0.19872560650110244,
      "num_tokens": 86008853.0,
      "step": 41440
    },
    {
      "entropy": 5.44201807975769,
      "epoch": 3.759524673439768,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003641866059762223,
      "loss": 5.0519,
      "mean_token_accuracy": 0.20527777224779128,
      "num_tokens": 86020358.0,
      "step": 41445
    },
    {
      "entropy": 5.501074647903442,
      "epoch": 3.759978229317852,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003641568991292747,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19790804833173753,
      "num_tokens": 86030706.0,
      "step": 41450
    },
    {
      "entropy": 5.6079573154449465,
      "epoch": 3.7604317851959363,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036412719043853956,
      "loss": 5.1176,
      "mean_token_accuracy": 0.20213967114686965,
      "num_tokens": 86041964.0,
      "step": 41455
    },
    {
      "entropy": 5.521636438369751,
      "epoch": 3.76088534107402,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036409747990463146,
      "loss": 5.0815,
      "mean_token_accuracy": 0.2099553406238556,
      "num_tokens": 86052213.0,
      "step": 41460
    },
    {
      "entropy": 5.385048532485962,
      "epoch": 3.7613388969521044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003640677675281649,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20008568614721298,
      "num_tokens": 86061419.0,
      "step": 41465
    },
    {
      "entropy": 5.519166088104248,
      "epoch": 3.7617924528301887,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003640380533097541,
      "loss": 5.0598,
      "mean_token_accuracy": 0.2039426878094673,
      "num_tokens": 86071585.0,
      "step": 41470
    },
    {
      "entropy": 5.529179668426513,
      "epoch": 3.762246008708273,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036400833725001364,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20105170458555222,
      "num_tokens": 86081175.0,
      "step": 41475
    },
    {
      "entropy": 5.499562931060791,
      "epoch": 3.762699564586357,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003639786193495581,
      "loss": 5.0916,
      "mean_token_accuracy": 0.20077542066574097,
      "num_tokens": 86092407.0,
      "step": 41480
    },
    {
      "entropy": 5.5628156661987305,
      "epoch": 3.763153120464441,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036394889960900215,
      "loss": 5.0655,
      "mean_token_accuracy": 0.20430233031511308,
      "num_tokens": 86102218.0,
      "step": 41485
    },
    {
      "entropy": 5.578890657424926,
      "epoch": 3.7636066763425253,
      "grad_norm": 1.0,
      "learning_rate": 0.0003639191780289602,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20714971274137498,
      "num_tokens": 86112631.0,
      "step": 41490
    },
    {
      "entropy": 5.488563442230225,
      "epoch": 3.7640602322206096,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0003638894546100469,
      "loss": 5.1191,
      "mean_token_accuracy": 0.20011247843503951,
      "num_tokens": 86123675.0,
      "step": 41495
    },
    {
      "entropy": 5.433043527603149,
      "epoch": 3.764513788098694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036385972935287715,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20734616369009018,
      "num_tokens": 86133909.0,
      "step": 41500
    },
    {
      "entropy": 5.520180749893188,
      "epoch": 3.764967343976778,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003638300022580656,
      "loss": 5.0863,
      "mean_token_accuracy": 0.20269207805395126,
      "num_tokens": 86144560.0,
      "step": 41505
    },
    {
      "entropy": 5.582522201538086,
      "epoch": 3.7654208998548624,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036380027332622684,
      "loss": 5.1025,
      "mean_token_accuracy": 0.19804415702819825,
      "num_tokens": 86155140.0,
      "step": 41510
    },
    {
      "entropy": 5.537507438659668,
      "epoch": 3.765874455732946,
      "grad_norm": 1.125,
      "learning_rate": 0.00036377054255797584,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19976139217615127,
      "num_tokens": 86165555.0,
      "step": 41515
    },
    {
      "entropy": 5.507036876678467,
      "epoch": 3.7663280116110305,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036374080995392737,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20509132891893386,
      "num_tokens": 86176204.0,
      "step": 41520
    },
    {
      "entropy": 5.536675214767456,
      "epoch": 3.7667815674891147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003637110755146964,
      "loss": 5.1347,
      "mean_token_accuracy": 0.2025540441274643,
      "num_tokens": 86186888.0,
      "step": 41525
    },
    {
      "entropy": 5.535649919509888,
      "epoch": 3.7672351233671986,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036368133924089766,
      "loss": 5.003,
      "mean_token_accuracy": 0.21679761707782746,
      "num_tokens": 86196654.0,
      "step": 41530
    },
    {
      "entropy": 5.4687753200531,
      "epoch": 3.767688679245283,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003636516011331464,
      "loss": 5.0094,
      "mean_token_accuracy": 0.212778140604496,
      "num_tokens": 86207045.0,
      "step": 41535
    },
    {
      "entropy": 5.498135900497436,
      "epoch": 3.768142235123367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036362186119205734,
      "loss": 5.102,
      "mean_token_accuracy": 0.20019730925559998,
      "num_tokens": 86216906.0,
      "step": 41540
    },
    {
      "entropy": 5.474241304397583,
      "epoch": 3.7685957910014514,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003635921194182457,
      "loss": 5.0428,
      "mean_token_accuracy": 0.2072534054517746,
      "num_tokens": 86227380.0,
      "step": 41545
    },
    {
      "entropy": 5.544825267791748,
      "epoch": 3.7690493468795356,
      "grad_norm": 1.0,
      "learning_rate": 0.0003635623758123263,
      "loss": 5.1842,
      "mean_token_accuracy": 0.20483894646167755,
      "num_tokens": 86236948.0,
      "step": 41550
    },
    {
      "entropy": 5.519665670394898,
      "epoch": 3.76950290275762,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036353263037491445,
      "loss": 5.1186,
      "mean_token_accuracy": 0.20025717467069626,
      "num_tokens": 86246784.0,
      "step": 41555
    },
    {
      "entropy": 5.43360857963562,
      "epoch": 3.769956458635704,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003635028831066253,
      "loss": 4.9863,
      "mean_token_accuracy": 0.20919818580150604,
      "num_tokens": 86257381.0,
      "step": 41560
    },
    {
      "entropy": 5.437063932418823,
      "epoch": 3.770410014513788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003634731340080739,
      "loss": 4.9975,
      "mean_token_accuracy": 0.2083994373679161,
      "num_tokens": 86268117.0,
      "step": 41565
    },
    {
      "entropy": 5.5972551822662355,
      "epoch": 3.7708635703918723,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003634433830798756,
      "loss": 5.1842,
      "mean_token_accuracy": 0.1993602082133293,
      "num_tokens": 86279195.0,
      "step": 41570
    },
    {
      "entropy": 5.384597826004028,
      "epoch": 3.7713171262699565,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003634136303226457,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22348787635564804,
      "num_tokens": 86288154.0,
      "step": 41575
    },
    {
      "entropy": 5.477279615402222,
      "epoch": 3.771770682148041,
      "grad_norm": 1.0,
      "learning_rate": 0.0003633838757369992,
      "loss": 5.0636,
      "mean_token_accuracy": 0.20130899399518967,
      "num_tokens": 86297524.0,
      "step": 41580
    },
    {
      "entropy": 5.497535991668701,
      "epoch": 3.7722242380261246,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003633541193235517,
      "loss": 5.1313,
      "mean_token_accuracy": 0.20162007957696915,
      "num_tokens": 86307031.0,
      "step": 41585
    },
    {
      "entropy": 5.47582368850708,
      "epoch": 3.772677793904209,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003633243610829185,
      "loss": 5.1082,
      "mean_token_accuracy": 0.2077782154083252,
      "num_tokens": 86317629.0,
      "step": 41590
    },
    {
      "entropy": 5.521573448181153,
      "epoch": 3.773131349782293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036329460101571496,
      "loss": 5.1159,
      "mean_token_accuracy": 0.20621962398290633,
      "num_tokens": 86328243.0,
      "step": 41595
    },
    {
      "entropy": 5.619073820114136,
      "epoch": 3.7735849056603774,
      "grad_norm": 1.0,
      "learning_rate": 0.0003632648391225568,
      "loss": 5.1583,
      "mean_token_accuracy": 0.20177863985300065,
      "num_tokens": 86338905.0,
      "step": 41600
    },
    {
      "entropy": 5.521846103668213,
      "epoch": 3.7740384615384617,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00036323507540405903,
      "loss": 5.0849,
      "mean_token_accuracy": 0.2037772938609123,
      "num_tokens": 86350433.0,
      "step": 41605
    },
    {
      "entropy": 5.52441897392273,
      "epoch": 3.774492017416546,
      "grad_norm": 1.125,
      "learning_rate": 0.0003632053098608374,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19655869752168656,
      "num_tokens": 86360664.0,
      "step": 41610
    },
    {
      "entropy": 5.542790269851684,
      "epoch": 3.7749455732946298,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036317554249350767,
      "loss": 5.1427,
      "mean_token_accuracy": 0.20174528807401657,
      "num_tokens": 86372101.0,
      "step": 41615
    },
    {
      "entropy": 5.507324266433716,
      "epoch": 3.775399129172714,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003631457733026851,
      "loss": 4.969,
      "mean_token_accuracy": 0.2096049815416336,
      "num_tokens": 86382494.0,
      "step": 41620
    },
    {
      "entropy": 5.607592582702637,
      "epoch": 3.7758526850507983,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036311600228898554,
      "loss": 5.2297,
      "mean_token_accuracy": 0.19313772022724152,
      "num_tokens": 86393292.0,
      "step": 41625
    },
    {
      "entropy": 5.425015783309936,
      "epoch": 3.7763062409288826,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036308622945302455,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21305023431777953,
      "num_tokens": 86402940.0,
      "step": 41630
    },
    {
      "entropy": 5.434909152984619,
      "epoch": 3.7767597968069664,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000363056454795418,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20755904465913771,
      "num_tokens": 86413382.0,
      "step": 41635
    },
    {
      "entropy": 5.562140083312988,
      "epoch": 3.7772133526850507,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003630266783167814,
      "loss": 5.1269,
      "mean_token_accuracy": 0.2010779470205307,
      "num_tokens": 86423866.0,
      "step": 41640
    },
    {
      "entropy": 5.6019398212432865,
      "epoch": 3.777666908563135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003629969000177307,
      "loss": 5.1804,
      "mean_token_accuracy": 0.19969211518764496,
      "num_tokens": 86434908.0,
      "step": 41645
    },
    {
      "entropy": 5.404290580749512,
      "epoch": 3.778120464441219,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003629671198988817,
      "loss": 4.9651,
      "mean_token_accuracy": 0.2201191157102585,
      "num_tokens": 86445604.0,
      "step": 41650
    },
    {
      "entropy": 5.427257394790649,
      "epoch": 3.7785740203193035,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003629373379608502,
      "loss": 4.9491,
      "mean_token_accuracy": 0.2146006017923355,
      "num_tokens": 86455813.0,
      "step": 41655
    },
    {
      "entropy": 5.404680442810059,
      "epoch": 3.7790275761973877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036290755420425214,
      "loss": 4.908,
      "mean_token_accuracy": 0.219795922935009,
      "num_tokens": 86465256.0,
      "step": 41660
    },
    {
      "entropy": 5.457080411911011,
      "epoch": 3.7794811320754715,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036287776862970356,
      "loss": 5.0262,
      "mean_token_accuracy": 0.21063854098320006,
      "num_tokens": 86474605.0,
      "step": 41665
    },
    {
      "entropy": 5.541619825363159,
      "epoch": 3.779934687953556,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036284798123782024,
      "loss": 5.0592,
      "mean_token_accuracy": 0.20611635148525237,
      "num_tokens": 86484886.0,
      "step": 41670
    },
    {
      "entropy": 5.502658891677856,
      "epoch": 3.78038824383164,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003628181920292182,
      "loss": 5.1297,
      "mean_token_accuracy": 0.19855596870183945,
      "num_tokens": 86495816.0,
      "step": 41675
    },
    {
      "entropy": 5.499367809295654,
      "epoch": 3.7808417997097243,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003627884010045138,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20667518377304078,
      "num_tokens": 86506319.0,
      "step": 41680
    },
    {
      "entropy": 5.562135839462281,
      "epoch": 3.781295355587808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003627586081643227,
      "loss": 5.0595,
      "mean_token_accuracy": 0.2056710258126259,
      "num_tokens": 86516952.0,
      "step": 41685
    },
    {
      "entropy": 5.512082242965699,
      "epoch": 3.7817489114658924,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036272881350926135,
      "loss": 5.1545,
      "mean_token_accuracy": 0.20294905453920364,
      "num_tokens": 86527327.0,
      "step": 41690
    },
    {
      "entropy": 5.497145509719848,
      "epoch": 3.7822024673439767,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003626990170399457,
      "loss": 5.0167,
      "mean_token_accuracy": 0.21135056912899017,
      "num_tokens": 86538281.0,
      "step": 41695
    },
    {
      "entropy": 5.513243007659912,
      "epoch": 3.782656023222061,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036266921875699205,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19860897660255433,
      "num_tokens": 86549034.0,
      "step": 41700
    },
    {
      "entropy": 5.530057621002197,
      "epoch": 3.7831095791001452,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003626394186610166,
      "loss": 5.0911,
      "mean_token_accuracy": 0.20470528304576874,
      "num_tokens": 86559306.0,
      "step": 41705
    },
    {
      "entropy": 5.581361484527588,
      "epoch": 3.7835631349782295,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003626096167526356,
      "loss": 5.0852,
      "mean_token_accuracy": 0.20216915458440782,
      "num_tokens": 86569414.0,
      "step": 41710
    },
    {
      "entropy": 5.378810024261474,
      "epoch": 3.7840166908563138,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003625798130324655,
      "loss": 4.914,
      "mean_token_accuracy": 0.21569414138793946,
      "num_tokens": 86578554.0,
      "step": 41715
    },
    {
      "entropy": 5.495085620880127,
      "epoch": 3.7844702467343976,
      "grad_norm": 1.125,
      "learning_rate": 0.00036255000750112256,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20675849616527558,
      "num_tokens": 86588624.0,
      "step": 41720
    },
    {
      "entropy": 5.495821285247803,
      "epoch": 3.784923802612482,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036252020015922304,
      "loss": 5.0813,
      "mean_token_accuracy": 0.20602733939886092,
      "num_tokens": 86599266.0,
      "step": 41725
    },
    {
      "entropy": 5.557767009735107,
      "epoch": 3.785377358490566,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036249039100738357,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1957371011376381,
      "num_tokens": 86609853.0,
      "step": 41730
    },
    {
      "entropy": 5.425660896301269,
      "epoch": 3.7858309143686504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003624605800462205,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20633552074432374,
      "num_tokens": 86619927.0,
      "step": 41735
    },
    {
      "entropy": 5.553432512283325,
      "epoch": 3.786284470246734,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036243076727635027,
      "loss": 5.0834,
      "mean_token_accuracy": 0.2082894414663315,
      "num_tokens": 86630079.0,
      "step": 41740
    },
    {
      "entropy": 5.490221261978149,
      "epoch": 3.7867380261248185,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036240095269838955,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20914474874734879,
      "num_tokens": 86639744.0,
      "step": 41745
    },
    {
      "entropy": 5.435925912857056,
      "epoch": 3.7871915820029027,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036237113631295485,
      "loss": 5.1026,
      "mean_token_accuracy": 0.19815556854009628,
      "num_tokens": 86650866.0,
      "step": 41750
    },
    {
      "entropy": 5.508586406707764,
      "epoch": 3.787645137880987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036234131812066286,
      "loss": 4.9397,
      "mean_token_accuracy": 0.21644827127456664,
      "num_tokens": 86659864.0,
      "step": 41755
    },
    {
      "entropy": 5.42819242477417,
      "epoch": 3.7880986937590713,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036231149812213006,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2067960575222969,
      "num_tokens": 86670039.0,
      "step": 41760
    },
    {
      "entropy": 5.440286302566529,
      "epoch": 3.7885522496371555,
      "grad_norm": 1.0,
      "learning_rate": 0.0003622816763179732,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20864178389310836,
      "num_tokens": 86680434.0,
      "step": 41765
    },
    {
      "entropy": 5.545727157592774,
      "epoch": 3.7890058055152394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003622518527088091,
      "loss": 5.0624,
      "mean_token_accuracy": 0.2026737079024315,
      "num_tokens": 86691026.0,
      "step": 41770
    },
    {
      "entropy": 5.550058364868164,
      "epoch": 3.7894593613933236,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036222202729525453,
      "loss": 5.105,
      "mean_token_accuracy": 0.20838313549757004,
      "num_tokens": 86701323.0,
      "step": 41775
    },
    {
      "entropy": 5.4153186798095705,
      "epoch": 3.789912917271408,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000362192200077926,
      "loss": 5.0071,
      "mean_token_accuracy": 0.21493927389383316,
      "num_tokens": 86711152.0,
      "step": 41780
    },
    {
      "entropy": 5.511436128616333,
      "epoch": 3.790366473149492,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036216237105744076,
      "loss": 5.1507,
      "mean_token_accuracy": 0.20446909964084625,
      "num_tokens": 86720814.0,
      "step": 41785
    },
    {
      "entropy": 5.5254209518432615,
      "epoch": 3.790820029027576,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003621325402344154,
      "loss": 5.0387,
      "mean_token_accuracy": 0.21204409450292588,
      "num_tokens": 86731229.0,
      "step": 41790
    },
    {
      "entropy": 5.493530035018921,
      "epoch": 3.7912735849056602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036210270760946684,
      "loss": 5.0103,
      "mean_token_accuracy": 0.2046894535422325,
      "num_tokens": 86741440.0,
      "step": 41795
    },
    {
      "entropy": 5.5528772354125975,
      "epoch": 3.7917271407837445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003620728731832122,
      "loss": 5.1538,
      "mean_token_accuracy": 0.19644194096326828,
      "num_tokens": 86752022.0,
      "step": 41800
    },
    {
      "entropy": 5.552502393722534,
      "epoch": 3.7921806966618288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003620430369562683,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20453015714883804,
      "num_tokens": 86763090.0,
      "step": 41805
    },
    {
      "entropy": 5.4454264640808105,
      "epoch": 3.792634252539913,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036201319892925214,
      "loss": 5.0215,
      "mean_token_accuracy": 0.20592166483402252,
      "num_tokens": 86773357.0,
      "step": 41810
    },
    {
      "entropy": 5.45047698020935,
      "epoch": 3.7930878084179973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000361983359102781,
      "loss": 4.9983,
      "mean_token_accuracy": 0.2025105983018875,
      "num_tokens": 86784402.0,
      "step": 41815
    },
    {
      "entropy": 5.531357955932617,
      "epoch": 3.793541364296081,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00036195351747747177,
      "loss": 5.1072,
      "mean_token_accuracy": 0.20053674429655075,
      "num_tokens": 86794945.0,
      "step": 41820
    },
    {
      "entropy": 5.457105302810669,
      "epoch": 3.7939949201741654,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036192367405394157,
      "loss": 5.0454,
      "mean_token_accuracy": 0.20321767330169677,
      "num_tokens": 86806936.0,
      "step": 41825
    },
    {
      "entropy": 5.562817668914795,
      "epoch": 3.7944484760522497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003618938288328077,
      "loss": 5.1535,
      "mean_token_accuracy": 0.2006813406944275,
      "num_tokens": 86817776.0,
      "step": 41830
    },
    {
      "entropy": 5.452915096282959,
      "epoch": 3.794902031930334,
      "grad_norm": 0.890625,
      "learning_rate": 0.0003618639818146873,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21324486285448074,
      "num_tokens": 86828054.0,
      "step": 41835
    },
    {
      "entropy": 5.481283473968506,
      "epoch": 3.7953555878084178,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00036183413300019767,
      "loss": 4.9944,
      "mean_token_accuracy": 0.21032946556806564,
      "num_tokens": 86838132.0,
      "step": 41840
    },
    {
      "entropy": 5.468667984008789,
      "epoch": 3.795809143686502,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036180428238995595,
      "loss": 5.047,
      "mean_token_accuracy": 0.2000681385397911,
      "num_tokens": 86848219.0,
      "step": 41845
    },
    {
      "entropy": 5.416669225692749,
      "epoch": 3.7962626995645863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003617744299845796,
      "loss": 4.9796,
      "mean_token_accuracy": 0.20830440372228623,
      "num_tokens": 86858559.0,
      "step": 41850
    },
    {
      "entropy": 5.4486891746521,
      "epoch": 3.7967162554426706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003617445757846859,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20766908675432205,
      "num_tokens": 86868573.0,
      "step": 41855
    },
    {
      "entropy": 5.565394353866577,
      "epoch": 3.797169811320755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036171471979089227,
      "loss": 5.0572,
      "mean_token_accuracy": 0.20579249858856202,
      "num_tokens": 86878641.0,
      "step": 41860
    },
    {
      "entropy": 5.539944839477539,
      "epoch": 3.797623367198839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003616848620038162,
      "loss": 5.091,
      "mean_token_accuracy": 0.20178698152303695,
      "num_tokens": 86888644.0,
      "step": 41865
    },
    {
      "entropy": 5.561775016784668,
      "epoch": 3.7980769230769234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000361655002424075,
      "loss": 5.1993,
      "mean_token_accuracy": 0.19604225009679793,
      "num_tokens": 86899428.0,
      "step": 41870
    },
    {
      "entropy": 5.542341566085815,
      "epoch": 3.798530478955007,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036162514105228635,
      "loss": 5.0699,
      "mean_token_accuracy": 0.20290592908859253,
      "num_tokens": 86910155.0,
      "step": 41875
    },
    {
      "entropy": 5.474612998962402,
      "epoch": 3.7989840348330914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003615952778890676,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21019109785556794,
      "num_tokens": 86919783.0,
      "step": 41880
    },
    {
      "entropy": 5.511729192733765,
      "epoch": 3.7994375907111757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036156541293503653,
      "loss": 5.0785,
      "mean_token_accuracy": 0.20306698828935624,
      "num_tokens": 86929647.0,
      "step": 41885
    },
    {
      "entropy": 5.480734825134277,
      "epoch": 3.7998911465892595,
      "grad_norm": 0.921875,
      "learning_rate": 0.00036153554619081064,
      "loss": 4.9949,
      "mean_token_accuracy": 0.20311630070209502,
      "num_tokens": 86940113.0,
      "step": 41890
    },
    {
      "entropy": 5.53898286819458,
      "epoch": 3.800344702467344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003615056776570076,
      "loss": 5.1164,
      "mean_token_accuracy": 0.20700255334377288,
      "num_tokens": 86949950.0,
      "step": 41895
    },
    {
      "entropy": 5.457999181747437,
      "epoch": 3.800798258345428,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036147580733424513,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21484729796648025,
      "num_tokens": 86959286.0,
      "step": 41900
    },
    {
      "entropy": 5.439472961425781,
      "epoch": 3.8012518142235123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003614459352231409,
      "loss": 5.0454,
      "mean_token_accuracy": 0.21074467450380324,
      "num_tokens": 86970165.0,
      "step": 41905
    },
    {
      "entropy": 5.459519529342652,
      "epoch": 3.8017053701015966,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003614160613243127,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20713070183992385,
      "num_tokens": 86980331.0,
      "step": 41910
    },
    {
      "entropy": 5.613748407363891,
      "epoch": 3.802158925979681,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003613861856383783,
      "loss": 5.1139,
      "mean_token_accuracy": 0.19770436584949494,
      "num_tokens": 86991790.0,
      "step": 41915
    },
    {
      "entropy": 5.560690975189209,
      "epoch": 3.802612481857765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003613563081659556,
      "loss": 5.1776,
      "mean_token_accuracy": 0.2075280725955963,
      "num_tokens": 87001220.0,
      "step": 41920
    },
    {
      "entropy": 5.532101774215699,
      "epoch": 3.803066037735849,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003613264289076625,
      "loss": 5.0545,
      "mean_token_accuracy": 0.2048826664686203,
      "num_tokens": 87012720.0,
      "step": 41925
    },
    {
      "entropy": 5.5116808891296385,
      "epoch": 3.803519593613933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003612965478641167,
      "loss": 5.0749,
      "mean_token_accuracy": 0.2049919545650482,
      "num_tokens": 87021891.0,
      "step": 41930
    },
    {
      "entropy": 5.595911931991577,
      "epoch": 3.8039731494920175,
      "grad_norm": 1.125,
      "learning_rate": 0.00036126666503593643,
      "loss": 5.1771,
      "mean_token_accuracy": 0.19516355991363527,
      "num_tokens": 87032546.0,
      "step": 41935
    },
    {
      "entropy": 5.538184404373169,
      "epoch": 3.8044267053701017,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003612367804237395,
      "loss": 4.9984,
      "mean_token_accuracy": 0.20233309864997864,
      "num_tokens": 87042173.0,
      "step": 41940
    },
    {
      "entropy": 5.539017152786255,
      "epoch": 3.8048802612481856,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003612068940281439,
      "loss": 5.1229,
      "mean_token_accuracy": 0.20611168295145035,
      "num_tokens": 87052475.0,
      "step": 41945
    },
    {
      "entropy": 5.489813423156738,
      "epoch": 3.80533381712627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003611770058497679,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20512388795614242,
      "num_tokens": 87062698.0,
      "step": 41950
    },
    {
      "entropy": 5.441302156448364,
      "epoch": 3.805787373004354,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036114711588922945,
      "loss": 5.0156,
      "mean_token_accuracy": 0.20422748178243638,
      "num_tokens": 87073168.0,
      "step": 41955
    },
    {
      "entropy": 5.553479957580566,
      "epoch": 3.8062409288824384,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003611172241471466,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19567430317401885,
      "num_tokens": 87084270.0,
      "step": 41960
    },
    {
      "entropy": 5.523167943954467,
      "epoch": 3.8066944847605226,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003610873306241376,
      "loss": 5.0703,
      "mean_token_accuracy": 0.20525579601526261,
      "num_tokens": 87093869.0,
      "step": 41965
    },
    {
      "entropy": 5.474940395355224,
      "epoch": 3.807148040638607,
      "grad_norm": 1.0,
      "learning_rate": 0.00036105743532082075,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2073134571313858,
      "num_tokens": 87103371.0,
      "step": 41970
    },
    {
      "entropy": 5.48955225944519,
      "epoch": 3.8076015965166907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003610275382378142,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20411363393068313,
      "num_tokens": 87113384.0,
      "step": 41975
    },
    {
      "entropy": 5.565797948837281,
      "epoch": 3.808055152394775,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003609976393757362,
      "loss": 5.1085,
      "mean_token_accuracy": 0.20280286818742752,
      "num_tokens": 87123925.0,
      "step": 41980
    },
    {
      "entropy": 5.512723207473755,
      "epoch": 3.8085087082728593,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036096773873520513,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2043600395321846,
      "num_tokens": 87134608.0,
      "step": 41985
    },
    {
      "entropy": 5.498331022262573,
      "epoch": 3.8089622641509435,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036093783631683927,
      "loss": 5.0333,
      "mean_token_accuracy": 0.208993861079216,
      "num_tokens": 87145184.0,
      "step": 41990
    },
    {
      "entropy": 5.449335145950317,
      "epoch": 3.8094158200290273,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003609079321212571,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21077121198177337,
      "num_tokens": 87155483.0,
      "step": 41995
    },
    {
      "entropy": 5.503931617736816,
      "epoch": 3.8098693759071116,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036087802614907696,
      "loss": 5.1341,
      "mean_token_accuracy": 0.19912753254175186,
      "num_tokens": 87166615.0,
      "step": 42000
    },
    {
      "epoch": 3.8098693759071116,
      "eval_entropy": 5.290621344320375,
      "eval_loss": 5.117408275604248,
      "eval_mean_token_accuracy": 0.2109953536074693,
      "eval_num_tokens": 87166615.0,
      "eval_runtime": 26.4342,
      "eval_samples_per_second": 1337.663,
      "eval_steps_per_second": 167.208,
      "step": 42000
    },
    {
      "entropy": 5.520386123657227,
      "epoch": 3.810322931785196,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036084811840091736,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20452774912118912,
      "num_tokens": 87177151.0,
      "step": 42005
    },
    {
      "entropy": 5.497348690032959,
      "epoch": 3.81077648766328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003608182088773969,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20773415714502336,
      "num_tokens": 87188085.0,
      "step": 42010
    },
    {
      "entropy": 5.450810384750366,
      "epoch": 3.8112300435413644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003607882975791338,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2071751669049263,
      "num_tokens": 87197240.0,
      "step": 42015
    },
    {
      "entropy": 5.517071056365967,
      "epoch": 3.8116835994194487,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000360758384506747,
      "loss": 5.0319,
      "mean_token_accuracy": 0.2123784229159355,
      "num_tokens": 87208064.0,
      "step": 42020
    },
    {
      "entropy": 5.473550701141358,
      "epoch": 3.8121371552975325,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036072846966085496,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21013074815273286,
      "num_tokens": 87218386.0,
      "step": 42025
    },
    {
      "entropy": 5.42082257270813,
      "epoch": 3.8125907111756168,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036069855304207634,
      "loss": 4.9381,
      "mean_token_accuracy": 0.21758072972297668,
      "num_tokens": 87228459.0,
      "step": 42030
    },
    {
      "entropy": 5.5243004322052,
      "epoch": 3.813044267053701,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003606686346510297,
      "loss": 5.1651,
      "mean_token_accuracy": 0.19441013783216476,
      "num_tokens": 87240317.0,
      "step": 42035
    },
    {
      "entropy": 5.419412136077881,
      "epoch": 3.8134978229317853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000360638714488334,
      "loss": 4.9392,
      "mean_token_accuracy": 0.21204402446746826,
      "num_tokens": 87250197.0,
      "step": 42040
    },
    {
      "entropy": 5.436420869827271,
      "epoch": 3.813951378809869,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036060879255460766,
      "loss": 5.0134,
      "mean_token_accuracy": 0.2079639971256256,
      "num_tokens": 87261266.0,
      "step": 42045
    },
    {
      "entropy": 5.4012500762939455,
      "epoch": 3.8144049346879534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003605788688504699,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2179838463664055,
      "num_tokens": 87270490.0,
      "step": 42050
    },
    {
      "entropy": 5.5060133934021,
      "epoch": 3.8148584905660377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003605489433765392,
      "loss": 5.0999,
      "mean_token_accuracy": 0.20217269361019136,
      "num_tokens": 87281260.0,
      "step": 42055
    },
    {
      "entropy": 5.427610158920288,
      "epoch": 3.815312046444122,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003605190161334345,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2152190923690796,
      "num_tokens": 87290577.0,
      "step": 42060
    },
    {
      "entropy": 5.400349903106689,
      "epoch": 3.815765602322206,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003604890871217748,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2177489772439003,
      "num_tokens": 87300185.0,
      "step": 42065
    },
    {
      "entropy": 5.535178709030151,
      "epoch": 3.8162191582002905,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003604591563421791,
      "loss": 5.1416,
      "mean_token_accuracy": 0.2037598505616188,
      "num_tokens": 87310088.0,
      "step": 42070
    },
    {
      "entropy": 5.5440315246582035,
      "epoch": 3.8166727140783747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003604292237952662,
      "loss": 5.0393,
      "mean_token_accuracy": 0.20462062060832978,
      "num_tokens": 87319942.0,
      "step": 42075
    },
    {
      "entropy": 5.510799074172974,
      "epoch": 3.8171262699564585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003603992894816551,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20745522379875184,
      "num_tokens": 87329175.0,
      "step": 42080
    },
    {
      "entropy": 5.406182146072387,
      "epoch": 3.817579825834543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000360369353401965,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20964243412017822,
      "num_tokens": 87338555.0,
      "step": 42085
    },
    {
      "entropy": 5.452936601638794,
      "epoch": 3.818033381712627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036033941555681486,
      "loss": 5.0231,
      "mean_token_accuracy": 0.21441818922758102,
      "num_tokens": 87348424.0,
      "step": 42090
    },
    {
      "entropy": 5.489368677139282,
      "epoch": 3.8184869375907113,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036030947594682395,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20759521126747132,
      "num_tokens": 87359724.0,
      "step": 42095
    },
    {
      "entropy": 5.526029205322265,
      "epoch": 3.818940493468795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003602795345726113,
      "loss": 5.137,
      "mean_token_accuracy": 0.20069431960582734,
      "num_tokens": 87371324.0,
      "step": 42100
    },
    {
      "entropy": 5.49879150390625,
      "epoch": 3.8193940493468794,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003602495914347962,
      "loss": 4.9552,
      "mean_token_accuracy": 0.2144071191549301,
      "num_tokens": 87380770.0,
      "step": 42105
    },
    {
      "entropy": 5.45381350517273,
      "epoch": 3.8198476052249637,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036021964653399773,
      "loss": 5.092,
      "mean_token_accuracy": 0.20186692029237746,
      "num_tokens": 87391123.0,
      "step": 42110
    },
    {
      "entropy": 5.494565010070801,
      "epoch": 3.820301161103048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036018969987083524,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20634487569332122,
      "num_tokens": 87401357.0,
      "step": 42115
    },
    {
      "entropy": 5.502290678024292,
      "epoch": 3.8207547169811322,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003601597514459281,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20732104927301406,
      "num_tokens": 87411373.0,
      "step": 42120
    },
    {
      "entropy": 5.461354494094849,
      "epoch": 3.8212082728592165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036012980125989567,
      "loss": 4.9886,
      "mean_token_accuracy": 0.21519655585289002,
      "num_tokens": 87421915.0,
      "step": 42125
    },
    {
      "entropy": 5.5133778095245365,
      "epoch": 3.8216618287373003,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003600998493133571,
      "loss": 5.0343,
      "mean_token_accuracy": 0.2072301223874092,
      "num_tokens": 87432148.0,
      "step": 42130
    },
    {
      "entropy": 5.490494108200073,
      "epoch": 3.8221153846153846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003600698956069321,
      "loss": 5.0617,
      "mean_token_accuracy": 0.20860881954431534,
      "num_tokens": 87442605.0,
      "step": 42135
    },
    {
      "entropy": 5.480601978302002,
      "epoch": 3.822568940493469,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003600399401412399,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20919538140296937,
      "num_tokens": 87452482.0,
      "step": 42140
    },
    {
      "entropy": 5.380608654022216,
      "epoch": 3.823022496371553,
      "grad_norm": 0.984375,
      "learning_rate": 0.00036000998291690005,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21686868518590927,
      "num_tokens": 87462074.0,
      "step": 42145
    },
    {
      "entropy": 5.533180093765258,
      "epoch": 3.823476052249637,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003599800239345321,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20650697946548463,
      "num_tokens": 87472377.0,
      "step": 42150
    },
    {
      "entropy": 5.487172842025757,
      "epoch": 3.823929608127721,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003599500631947557,
      "loss": 5.0512,
      "mean_token_accuracy": 0.2063886657357216,
      "num_tokens": 87483436.0,
      "step": 42155
    },
    {
      "entropy": 5.52595705986023,
      "epoch": 3.8243831640058055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035992010069819025,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21108095943927765,
      "num_tokens": 87492231.0,
      "step": 42160
    },
    {
      "entropy": 5.573286008834839,
      "epoch": 3.8248367198838897,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035989013644545555,
      "loss": 5.1316,
      "mean_token_accuracy": 0.20474890023469924,
      "num_tokens": 87503416.0,
      "step": 42165
    },
    {
      "entropy": 5.463798379898071,
      "epoch": 3.825290275761974,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035986017043717114,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20688357800245286,
      "num_tokens": 87513737.0,
      "step": 42170
    },
    {
      "entropy": 5.530929613113403,
      "epoch": 3.8257438316400583,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003598302026739568,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20592880696058274,
      "num_tokens": 87523532.0,
      "step": 42175
    },
    {
      "entropy": 5.466580390930176,
      "epoch": 3.826197387518142,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00035980023315643217,
      "loss": 4.9289,
      "mean_token_accuracy": 0.218870048224926,
      "num_tokens": 87534066.0,
      "step": 42180
    },
    {
      "entropy": 5.5557982444763185,
      "epoch": 3.8266509433962264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003597702618852172,
      "loss": 5.1687,
      "mean_token_accuracy": 0.19076651334762573,
      "num_tokens": 87544520.0,
      "step": 42185
    },
    {
      "entropy": 5.4604576587677,
      "epoch": 3.8271044992743106,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003597402888609316,
      "loss": 5.0457,
      "mean_token_accuracy": 0.2109432339668274,
      "num_tokens": 87553913.0,
      "step": 42190
    },
    {
      "entropy": 5.532728815078736,
      "epoch": 3.827558055152395,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035971031408419524,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2072523608803749,
      "num_tokens": 87563434.0,
      "step": 42195
    },
    {
      "entropy": 5.5133625030517575,
      "epoch": 3.8280116110304787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035968033755562795,
      "loss": 5.065,
      "mean_token_accuracy": 0.2088916391134262,
      "num_tokens": 87573137.0,
      "step": 42200
    },
    {
      "entropy": 5.538109064102173,
      "epoch": 3.828465166908563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003596503592758497,
      "loss": 5.0591,
      "mean_token_accuracy": 0.20599123686552048,
      "num_tokens": 87582610.0,
      "step": 42205
    },
    {
      "entropy": 5.473601293563843,
      "epoch": 3.8289187227866472,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003596203792454805,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20961015820503234,
      "num_tokens": 87593062.0,
      "step": 42210
    },
    {
      "entropy": 5.485382127761841,
      "epoch": 3.8293722786647315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003595903974651403,
      "loss": 5.108,
      "mean_token_accuracy": 0.2030020460486412,
      "num_tokens": 87604250.0,
      "step": 42215
    },
    {
      "entropy": 5.5571025848388675,
      "epoch": 3.8298258345428158,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003595604139354491,
      "loss": 5.0961,
      "mean_token_accuracy": 0.20357718020677568,
      "num_tokens": 87615434.0,
      "step": 42220
    },
    {
      "entropy": 5.526327276229859,
      "epoch": 3.8302793904209,
      "grad_norm": 1.09375,
      "learning_rate": 0.000359530428657027,
      "loss": 5.0309,
      "mean_token_accuracy": 0.21372532695531846,
      "num_tokens": 87625005.0,
      "step": 42225
    },
    {
      "entropy": 5.465586614608765,
      "epoch": 3.8307329462989843,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003595004416304941,
      "loss": 4.9956,
      "mean_token_accuracy": 0.2084898054599762,
      "num_tokens": 87635486.0,
      "step": 42230
    },
    {
      "entropy": 5.489100551605224,
      "epoch": 3.831186502177068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035947045285647045,
      "loss": 5.0468,
      "mean_token_accuracy": 0.2063521161675453,
      "num_tokens": 87645771.0,
      "step": 42235
    },
    {
      "entropy": 5.5927605628967285,
      "epoch": 3.8316400580551524,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003594404623355765,
      "loss": 5.0979,
      "mean_token_accuracy": 0.20646103769540786,
      "num_tokens": 87656116.0,
      "step": 42240
    },
    {
      "entropy": 5.629701280593872,
      "epoch": 3.8320936139332367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003594104700684321,
      "loss": 5.2448,
      "mean_token_accuracy": 0.1977214440703392,
      "num_tokens": 87666476.0,
      "step": 42245
    },
    {
      "entropy": 5.469891500473023,
      "epoch": 3.8325471698113205,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035938047605565764,
      "loss": 5.056,
      "mean_token_accuracy": 0.2141733318567276,
      "num_tokens": 87677406.0,
      "step": 42250
    },
    {
      "entropy": 5.466805458068848,
      "epoch": 3.8330007256894048,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035935048029787356,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21631811708211898,
      "num_tokens": 87687461.0,
      "step": 42255
    },
    {
      "entropy": 5.474397897720337,
      "epoch": 3.833454281567489,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003593204827957,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20623654574155809,
      "num_tokens": 87696470.0,
      "step": 42260
    },
    {
      "entropy": 5.4830286502838135,
      "epoch": 3.8339078374455733,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003592904835497574,
      "loss": 5.0826,
      "mean_token_accuracy": 0.20478060096502304,
      "num_tokens": 87706598.0,
      "step": 42265
    },
    {
      "entropy": 5.582409143447876,
      "epoch": 3.8343613933236576,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003592604825606661,
      "loss": 5.1319,
      "mean_token_accuracy": 0.20215307474136351,
      "num_tokens": 87717749.0,
      "step": 42270
    },
    {
      "entropy": 5.47212700843811,
      "epoch": 3.834814949201742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003592304798290465,
      "loss": 4.9814,
      "mean_token_accuracy": 0.21388375014066696,
      "num_tokens": 87727893.0,
      "step": 42275
    },
    {
      "entropy": 5.41770453453064,
      "epoch": 3.835268505079826,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003592004753555191,
      "loss": 5.0652,
      "mean_token_accuracy": 0.20617363303899766,
      "num_tokens": 87737980.0,
      "step": 42280
    },
    {
      "entropy": 5.467943954467773,
      "epoch": 3.83572206095791,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003591704691407045,
      "loss": 5.0361,
      "mean_token_accuracy": 0.2092646598815918,
      "num_tokens": 87749271.0,
      "step": 42285
    },
    {
      "entropy": 5.4732390403747555,
      "epoch": 3.836175616835994,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035914046118522313,
      "loss": 5.0172,
      "mean_token_accuracy": 0.21270287334918975,
      "num_tokens": 87760523.0,
      "step": 42290
    },
    {
      "entropy": 5.466706037521362,
      "epoch": 3.8366291727140784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035911045148969556,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20953099727630614,
      "num_tokens": 87770084.0,
      "step": 42295
    },
    {
      "entropy": 5.574228668212891,
      "epoch": 3.8370827285921627,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00035908044005474244,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19903675466775894,
      "num_tokens": 87780163.0,
      "step": 42300
    },
    {
      "entropy": 5.4114298820495605,
      "epoch": 3.8375362844702465,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003590504268809843,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21158635914325713,
      "num_tokens": 87791382.0,
      "step": 42305
    },
    {
      "entropy": 5.374681043624878,
      "epoch": 3.837989840348331,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035902041196904204,
      "loss": 4.9886,
      "mean_token_accuracy": 0.2084943026304245,
      "num_tokens": 87802143.0,
      "step": 42310
    },
    {
      "entropy": 5.506963777542114,
      "epoch": 3.838443396226415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003589903953195362,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20508768558502197,
      "num_tokens": 87812255.0,
      "step": 42315
    },
    {
      "entropy": 5.550964784622193,
      "epoch": 3.8388969521044993,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035896037693308757,
      "loss": 5.0489,
      "mean_token_accuracy": 0.20595334768295287,
      "num_tokens": 87822744.0,
      "step": 42320
    },
    {
      "entropy": 5.568259191513062,
      "epoch": 3.8393505079825836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000358930356810317,
      "loss": 5.121,
      "mean_token_accuracy": 0.20090816766023636,
      "num_tokens": 87833701.0,
      "step": 42325
    },
    {
      "entropy": 5.528118085861206,
      "epoch": 3.839804063860668,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003589003349518452,
      "loss": 5.0814,
      "mean_token_accuracy": 0.20084428191184997,
      "num_tokens": 87844215.0,
      "step": 42330
    },
    {
      "entropy": 5.544552040100098,
      "epoch": 3.8402576197387517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035887031135829305,
      "loss": 5.065,
      "mean_token_accuracy": 0.20393575727939606,
      "num_tokens": 87854281.0,
      "step": 42335
    },
    {
      "entropy": 5.525538969039917,
      "epoch": 3.840711175616836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003588402860302816,
      "loss": 5.1283,
      "mean_token_accuracy": 0.20022645741701126,
      "num_tokens": 87866003.0,
      "step": 42340
    },
    {
      "entropy": 5.4943132400512695,
      "epoch": 3.84116473149492,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035881025896843165,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20673414766788484,
      "num_tokens": 87875854.0,
      "step": 42345
    },
    {
      "entropy": 5.548939514160156,
      "epoch": 3.8416182873730045,
      "grad_norm": 0.96875,
      "learning_rate": 0.00035878023017336415,
      "loss": 5.1635,
      "mean_token_accuracy": 0.20270679593086244,
      "num_tokens": 87887242.0,
      "step": 42350
    },
    {
      "entropy": 5.53283052444458,
      "epoch": 3.8420718432510883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003587501996457001,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2064540609717369,
      "num_tokens": 87897859.0,
      "step": 42355
    },
    {
      "entropy": 5.50273551940918,
      "epoch": 3.8425253991291726,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035872016738606067,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20755570083856584,
      "num_tokens": 87908945.0,
      "step": 42360
    },
    {
      "entropy": 5.452647113800049,
      "epoch": 3.842978955007257,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035869013339506674,
      "loss": 5.03,
      "mean_token_accuracy": 0.2098871424794197,
      "num_tokens": 87919572.0,
      "step": 42365
    },
    {
      "entropy": 5.549891948699951,
      "epoch": 3.843432510885341,
      "grad_norm": 1.125,
      "learning_rate": 0.00035866009767333964,
      "loss": 5.0897,
      "mean_token_accuracy": 0.20328256487846375,
      "num_tokens": 87929337.0,
      "step": 42370
    },
    {
      "entropy": 5.59000883102417,
      "epoch": 3.8438860667634254,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035863006022150034,
      "loss": 5.1395,
      "mean_token_accuracy": 0.20017314255237578,
      "num_tokens": 87940513.0,
      "step": 42375
    },
    {
      "entropy": 5.4650280475616455,
      "epoch": 3.8443396226415096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035860002104017014,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20965767800807952,
      "num_tokens": 87951181.0,
      "step": 42380
    },
    {
      "entropy": 5.49062147140503,
      "epoch": 3.8447931785195935,
      "grad_norm": 1.0,
      "learning_rate": 0.0003585699801299701,
      "loss": 5.0504,
      "mean_token_accuracy": 0.2068266674876213,
      "num_tokens": 87961925.0,
      "step": 42385
    },
    {
      "entropy": 5.475308609008789,
      "epoch": 3.8452467343976777,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035853993749152165,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20846969932317733,
      "num_tokens": 87972233.0,
      "step": 42390
    },
    {
      "entropy": 5.528082752227784,
      "epoch": 3.845700290275762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035850989312544596,
      "loss": 5.0952,
      "mean_token_accuracy": 0.20368697345256806,
      "num_tokens": 87983131.0,
      "step": 42395
    },
    {
      "entropy": 5.462475204467774,
      "epoch": 3.8461538461538463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003584798470323644,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21160053759813308,
      "num_tokens": 87993926.0,
      "step": 42400
    },
    {
      "entropy": 5.55931658744812,
      "epoch": 3.84660740203193,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003584497992128984,
      "loss": 5.1693,
      "mean_token_accuracy": 0.1962766796350479,
      "num_tokens": 88004403.0,
      "step": 42405
    },
    {
      "entropy": 5.496698951721191,
      "epoch": 3.8470609579100143,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003584197496676692,
      "loss": 5.0023,
      "mean_token_accuracy": 0.21319792419672012,
      "num_tokens": 88015289.0,
      "step": 42410
    },
    {
      "entropy": 5.432304096221924,
      "epoch": 3.8475145137880986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035838969839729846,
      "loss": 4.988,
      "mean_token_accuracy": 0.20772961378097535,
      "num_tokens": 88025724.0,
      "step": 42415
    },
    {
      "entropy": 5.485240697860718,
      "epoch": 3.847968069666183,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003583596454024073,
      "loss": 5.0983,
      "mean_token_accuracy": 0.20245155245065688,
      "num_tokens": 88036444.0,
      "step": 42420
    },
    {
      "entropy": 5.549129867553711,
      "epoch": 3.848421625544267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003583295906836176,
      "loss": 5.0874,
      "mean_token_accuracy": 0.2059298112988472,
      "num_tokens": 88047334.0,
      "step": 42425
    },
    {
      "entropy": 5.479011869430542,
      "epoch": 3.8488751814223514,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00035829953424155065,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20731521248817444,
      "num_tokens": 88057268.0,
      "step": 42430
    },
    {
      "entropy": 5.465181875228882,
      "epoch": 3.8493287373004357,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003582694760768281,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2090430200099945,
      "num_tokens": 88067615.0,
      "step": 42435
    },
    {
      "entropy": 5.530232381820679,
      "epoch": 3.8497822931785195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035823941619007155,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20371018946170807,
      "num_tokens": 88077754.0,
      "step": 42440
    },
    {
      "entropy": 5.476480960845947,
      "epoch": 3.8502358490566038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035820935458190275,
      "loss": 5.0297,
      "mean_token_accuracy": 0.2042573794722557,
      "num_tokens": 88088346.0,
      "step": 42445
    },
    {
      "entropy": 5.431892585754395,
      "epoch": 3.850689404934688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035817929125294323,
      "loss": 4.973,
      "mean_token_accuracy": 0.212597818672657,
      "num_tokens": 88098534.0,
      "step": 42450
    },
    {
      "entropy": 5.465068817138672,
      "epoch": 3.8511429608127723,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003581492262038148,
      "loss": 5.0516,
      "mean_token_accuracy": 0.2074795752763748,
      "num_tokens": 88109264.0,
      "step": 42455
    },
    {
      "entropy": 5.516505002975464,
      "epoch": 3.851596516690856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003581191594351392,
      "loss": 5.091,
      "mean_token_accuracy": 0.20537717640399933,
      "num_tokens": 88120407.0,
      "step": 42460
    },
    {
      "entropy": 5.532539319992066,
      "epoch": 3.8520500725689404,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003580890909475381,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20120948404073716,
      "num_tokens": 88131908.0,
      "step": 42465
    },
    {
      "entropy": 5.4513896942138675,
      "epoch": 3.8525036284470247,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035805902074163354,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20383096784353255,
      "num_tokens": 88142559.0,
      "step": 42470
    },
    {
      "entropy": 5.4637044906616214,
      "epoch": 3.852957184325109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003580289488180472,
      "loss": 5.0427,
      "mean_token_accuracy": 0.21050262749195098,
      "num_tokens": 88153441.0,
      "step": 42475
    },
    {
      "entropy": 5.492034387588501,
      "epoch": 3.853410740203193,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035799887517740103,
      "loss": 5.0659,
      "mean_token_accuracy": 0.209297838807106,
      "num_tokens": 88164153.0,
      "step": 42480
    },
    {
      "entropy": 5.551849746704102,
      "epoch": 3.8538642960812775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000357968799820317,
      "loss": 5.0503,
      "mean_token_accuracy": 0.2045384094119072,
      "num_tokens": 88174138.0,
      "step": 42485
    },
    {
      "entropy": 5.526755666732788,
      "epoch": 3.8543178519593613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003579387227474171,
      "loss": 5.0503,
      "mean_token_accuracy": 0.20492410212755202,
      "num_tokens": 88184704.0,
      "step": 42490
    },
    {
      "entropy": 5.550944042205811,
      "epoch": 3.8547714078374455,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003579086439593232,
      "loss": 5.0925,
      "mean_token_accuracy": 0.20356247276067735,
      "num_tokens": 88194225.0,
      "step": 42495
    },
    {
      "entropy": 5.518636274337768,
      "epoch": 3.85522496371553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003578785634566574,
      "loss": 5.0815,
      "mean_token_accuracy": 0.2023499935865402,
      "num_tokens": 88206125.0,
      "step": 42500
    },
    {
      "entropy": 5.432632875442505,
      "epoch": 3.855678519593614,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035784848124004187,
      "loss": 4.918,
      "mean_token_accuracy": 0.21862925440073014,
      "num_tokens": 88217085.0,
      "step": 42505
    },
    {
      "entropy": 5.465728998184204,
      "epoch": 3.856132075471698,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003578183973100986,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20457614809274674,
      "num_tokens": 88228126.0,
      "step": 42510
    },
    {
      "entropy": 5.521930265426636,
      "epoch": 3.856585631349782,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035778831166744966,
      "loss": 5.1454,
      "mean_token_accuracy": 0.1982669547200203,
      "num_tokens": 88238829.0,
      "step": 42515
    },
    {
      "entropy": 5.520575380325317,
      "epoch": 3.8570391872278664,
      "grad_norm": 1.125,
      "learning_rate": 0.0003577582243127175,
      "loss": 5.0614,
      "mean_token_accuracy": 0.20289280116558076,
      "num_tokens": 88249821.0,
      "step": 42520
    },
    {
      "entropy": 5.482039117813111,
      "epoch": 3.8574927431059507,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003577281352465241,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21038202494382857,
      "num_tokens": 88260126.0,
      "step": 42525
    },
    {
      "entropy": 5.476259183883667,
      "epoch": 3.857946298984035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035769804446949177,
      "loss": 5.0528,
      "mean_token_accuracy": 0.2092612400650978,
      "num_tokens": 88271492.0,
      "step": 42530
    },
    {
      "entropy": 5.566465044021607,
      "epoch": 3.8583998548621192,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003576679519822428,
      "loss": 5.1231,
      "mean_token_accuracy": 0.20039983093738556,
      "num_tokens": 88281966.0,
      "step": 42535
    },
    {
      "entropy": 5.401264286041259,
      "epoch": 3.858853410740203,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035763785778539963,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21533740311861038,
      "num_tokens": 88292805.0,
      "step": 42540
    },
    {
      "entropy": 5.468257427215576,
      "epoch": 3.8593069666182873,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003576077618795844,
      "loss": 4.9921,
      "mean_token_accuracy": 0.2079550266265869,
      "num_tokens": 88302727.0,
      "step": 42545
    },
    {
      "entropy": 5.401624441146851,
      "epoch": 3.8597605224963716,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003575776642654196,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20822807103395463,
      "num_tokens": 88313150.0,
      "step": 42550
    },
    {
      "entropy": 5.4548938274383545,
      "epoch": 3.860214078374456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035754756494352767,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21086847335100173,
      "num_tokens": 88323060.0,
      "step": 42555
    },
    {
      "entropy": 5.542991399765015,
      "epoch": 3.8606676342525397,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003575174639145311,
      "loss": 5.0234,
      "mean_token_accuracy": 0.21652293652296067,
      "num_tokens": 88332641.0,
      "step": 42560
    },
    {
      "entropy": 5.417452621459961,
      "epoch": 3.861121190130624,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003574873611790523,
      "loss": 5.0381,
      "mean_token_accuracy": 0.21209495216608049,
      "num_tokens": 88342807.0,
      "step": 42565
    },
    {
      "entropy": 5.4036603450775145,
      "epoch": 3.861574746008708,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035745725673771387,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21663809567689896,
      "num_tokens": 88352958.0,
      "step": 42570
    },
    {
      "entropy": 5.558398914337158,
      "epoch": 3.8620283018867925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035742715059113836,
      "loss": 5.0714,
      "mean_token_accuracy": 0.1988092079758644,
      "num_tokens": 88363184.0,
      "step": 42575
    },
    {
      "entropy": 5.402445554733276,
      "epoch": 3.8624818577648767,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003573970427399484,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2145727813243866,
      "num_tokens": 88373991.0,
      "step": 42580
    },
    {
      "entropy": 5.4906933307647705,
      "epoch": 3.862935413642961,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035736693318476646,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20903784036636353,
      "num_tokens": 88384518.0,
      "step": 42585
    },
    {
      "entropy": 5.3960693836212155,
      "epoch": 3.8633889695210453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035733682192621555,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2151514157652855,
      "num_tokens": 88394506.0,
      "step": 42590
    },
    {
      "entropy": 5.534240436553955,
      "epoch": 3.863842525399129,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003573067089649181,
      "loss": 5.0345,
      "mean_token_accuracy": 0.2058042511343956,
      "num_tokens": 88404625.0,
      "step": 42595
    },
    {
      "entropy": 5.530741119384766,
      "epoch": 3.8642960812772134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035727659430149686,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20377998650074006,
      "num_tokens": 88413754.0,
      "step": 42600
    },
    {
      "entropy": 5.529934883117676,
      "epoch": 3.8647496371552976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003572464779365748,
      "loss": 5.1288,
      "mean_token_accuracy": 0.19939050376415252,
      "num_tokens": 88424613.0,
      "step": 42605
    },
    {
      "entropy": 5.520506954193115,
      "epoch": 3.8652031930333814,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035721635987077455,
      "loss": 5.0491,
      "mean_token_accuracy": 0.2084427922964096,
      "num_tokens": 88435248.0,
      "step": 42610
    },
    {
      "entropy": 5.453482723236084,
      "epoch": 3.8656567489114657,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035718624010471904,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22097446620464326,
      "num_tokens": 88444802.0,
      "step": 42615
    },
    {
      "entropy": 5.45304536819458,
      "epoch": 3.86611030478955,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003571561186390311,
      "loss": 5.0691,
      "mean_token_accuracy": 0.20259811729192734,
      "num_tokens": 88455335.0,
      "step": 42620
    },
    {
      "entropy": 5.530066204071045,
      "epoch": 3.8665638606676342,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003571259954743338,
      "loss": 5.075,
      "mean_token_accuracy": 0.2058267518877983,
      "num_tokens": 88465857.0,
      "step": 42625
    },
    {
      "entropy": 5.499158477783203,
      "epoch": 3.8670174165457185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035709587061124985,
      "loss": 5.0978,
      "mean_token_accuracy": 0.2035111278295517,
      "num_tokens": 88475930.0,
      "step": 42630
    },
    {
      "entropy": 5.467127656936645,
      "epoch": 3.8674709724238028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035706574405040245,
      "loss": 5.0403,
      "mean_token_accuracy": 0.2121370479464531,
      "num_tokens": 88486732.0,
      "step": 42635
    },
    {
      "entropy": 5.514516162872314,
      "epoch": 3.867924528301887,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003570356157924145,
      "loss": 5.0428,
      "mean_token_accuracy": 0.20822187811136245,
      "num_tokens": 88497017.0,
      "step": 42640
    },
    {
      "entropy": 5.467503929138184,
      "epoch": 3.868378084179971,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003570054858379092,
      "loss": 4.9747,
      "mean_token_accuracy": 0.2186335578560829,
      "num_tokens": 88507423.0,
      "step": 42645
    },
    {
      "entropy": 5.486092472076416,
      "epoch": 3.868831640058055,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035697535418750955,
      "loss": 5.0241,
      "mean_token_accuracy": 0.2074723571538925,
      "num_tokens": 88517795.0,
      "step": 42650
    },
    {
      "entropy": 5.381175422668457,
      "epoch": 3.8692851959361394,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035694522084183864,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21356277614831926,
      "num_tokens": 88527641.0,
      "step": 42655
    },
    {
      "entropy": 5.477459335327149,
      "epoch": 3.8697387518142237,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035691508580151966,
      "loss": 5.0594,
      "mean_token_accuracy": 0.20150842517614365,
      "num_tokens": 88537874.0,
      "step": 42660
    },
    {
      "entropy": 5.507590961456299,
      "epoch": 3.8701923076923075,
      "grad_norm": 1.0,
      "learning_rate": 0.0003568849490671759,
      "loss": 5.042,
      "mean_token_accuracy": 0.209032142162323,
      "num_tokens": 88548108.0,
      "step": 42665
    },
    {
      "entropy": 5.407514905929565,
      "epoch": 3.8706458635703918,
      "grad_norm": 1.125,
      "learning_rate": 0.0003568548106394305,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2152765065431595,
      "num_tokens": 88558284.0,
      "step": 42670
    },
    {
      "entropy": 5.472148704528808,
      "epoch": 3.871099419448476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003568246705189067,
      "loss": 5.0305,
      "mean_token_accuracy": 0.20876606404781342,
      "num_tokens": 88568635.0,
      "step": 42675
    },
    {
      "entropy": 5.503427171707154,
      "epoch": 3.8715529753265603,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003567945287062279,
      "loss": 5.0415,
      "mean_token_accuracy": 0.20321287363767623,
      "num_tokens": 88577968.0,
      "step": 42680
    },
    {
      "entropy": 5.551769685745239,
      "epoch": 3.8720065312046446,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00035676438520201745,
      "loss": 5.1048,
      "mean_token_accuracy": 0.20441154390573502,
      "num_tokens": 88590562.0,
      "step": 42685
    },
    {
      "entropy": 5.506048154830933,
      "epoch": 3.872460087082729,
      "grad_norm": 0.9375,
      "learning_rate": 0.00035673424000689857,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20019052028656006,
      "num_tokens": 88601187.0,
      "step": 42690
    },
    {
      "entropy": 5.415367078781128,
      "epoch": 3.8729136429608126,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035670409312149495,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20771615207195282,
      "num_tokens": 88611459.0,
      "step": 42695
    },
    {
      "entropy": 5.480759763717652,
      "epoch": 3.873367198838897,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035667394454642963,
      "loss": 5.0696,
      "mean_token_accuracy": 0.2119443744421005,
      "num_tokens": 88621652.0,
      "step": 42700
    },
    {
      "entropy": 5.457339096069336,
      "epoch": 3.873820754716981,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003566437942823265,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21016415506601333,
      "num_tokens": 88632425.0,
      "step": 42705
    },
    {
      "entropy": 5.420820856094361,
      "epoch": 3.8742743105950654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035661364232980884,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22161610573530197,
      "num_tokens": 88641263.0,
      "step": 42710
    },
    {
      "entropy": 5.447889614105224,
      "epoch": 3.8747278664731493,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035658348868950027,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20393438041210174,
      "num_tokens": 88651665.0,
      "step": 42715
    },
    {
      "entropy": 5.431807565689087,
      "epoch": 3.8751814223512335,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035655333336202436,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21570306718349458,
      "num_tokens": 88660791.0,
      "step": 42720
    },
    {
      "entropy": 5.401371812820434,
      "epoch": 3.875634978229318,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003565231763480047,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21266758292913437,
      "num_tokens": 88671341.0,
      "step": 42725
    },
    {
      "entropy": 5.507083320617676,
      "epoch": 3.876088534107402,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003564930176480651,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19884066730737687,
      "num_tokens": 88682500.0,
      "step": 42730
    },
    {
      "entropy": 5.525128793716431,
      "epoch": 3.8765420899854863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035646285726282896,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20319928079843522,
      "num_tokens": 88693513.0,
      "step": 42735
    },
    {
      "entropy": 5.514414691925049,
      "epoch": 3.8769956458635706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035643269519292027,
      "loss": 5.1091,
      "mean_token_accuracy": 0.2054834857583046,
      "num_tokens": 88703498.0,
      "step": 42740
    },
    {
      "entropy": 5.445942401885986,
      "epoch": 3.8774492017416544,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003564025314389627,
      "loss": 5.0153,
      "mean_token_accuracy": 0.2046381115913391,
      "num_tokens": 88714217.0,
      "step": 42745
    },
    {
      "entropy": 5.489507865905762,
      "epoch": 3.8779027576197387,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00035637236600157994,
      "loss": 5.0595,
      "mean_token_accuracy": 0.2043559581041336,
      "num_tokens": 88725006.0,
      "step": 42750
    },
    {
      "entropy": 5.429996109008789,
      "epoch": 3.878356313497823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000356342198881396,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21702402979135513,
      "num_tokens": 88735253.0,
      "step": 42755
    },
    {
      "entropy": 5.50838942527771,
      "epoch": 3.878809869375907,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00035631203007903465,
      "loss": 5.0928,
      "mean_token_accuracy": 0.20458886176347732,
      "num_tokens": 88746313.0,
      "step": 42760
    },
    {
      "entropy": 5.472730112075806,
      "epoch": 3.879263425253991,
      "grad_norm": 1.0,
      "learning_rate": 0.0003562818595951198,
      "loss": 4.9892,
      "mean_token_accuracy": 0.2108807682991028,
      "num_tokens": 88757448.0,
      "step": 42765
    },
    {
      "entropy": 5.498257780075074,
      "epoch": 3.8797169811320753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003562516874302754,
      "loss": 5.0516,
      "mean_token_accuracy": 0.21314364075660705,
      "num_tokens": 88767426.0,
      "step": 42770
    },
    {
      "entropy": 5.501209259033203,
      "epoch": 3.8801705370101596,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003562215135851253,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20821274518966676,
      "num_tokens": 88778188.0,
      "step": 42775
    },
    {
      "entropy": 5.593484592437744,
      "epoch": 3.880624092888244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035619133806029374,
      "loss": 5.1116,
      "mean_token_accuracy": 0.2054688662290573,
      "num_tokens": 88788927.0,
      "step": 42780
    },
    {
      "entropy": 5.5225354671478275,
      "epoch": 3.881077648766328,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035616116085640455,
      "loss": 5.0738,
      "mean_token_accuracy": 0.20887231975793838,
      "num_tokens": 88798809.0,
      "step": 42785
    },
    {
      "entropy": 5.366028547286987,
      "epoch": 3.8815312046444124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003561309819740819,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22248044610023499,
      "num_tokens": 88808688.0,
      "step": 42790
    },
    {
      "entropy": 5.48792667388916,
      "epoch": 3.8819847605224966,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035610080141394983,
      "loss": 5.001,
      "mean_token_accuracy": 0.21254732012748717,
      "num_tokens": 88818400.0,
      "step": 42795
    },
    {
      "entropy": 5.440720558166504,
      "epoch": 3.8824383164005805,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035607061917663254,
      "loss": 5.0237,
      "mean_token_accuracy": 0.21002939939498902,
      "num_tokens": 88828032.0,
      "step": 42800
    },
    {
      "entropy": 5.492562675476075,
      "epoch": 3.8828918722786647,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035604043526275413,
      "loss": 5.1129,
      "mean_token_accuracy": 0.20692140907049178,
      "num_tokens": 88839856.0,
      "step": 42805
    },
    {
      "entropy": 5.540346002578735,
      "epoch": 3.883345428156749,
      "grad_norm": 1.0,
      "learning_rate": 0.000356010249672939,
      "loss": 5.0208,
      "mean_token_accuracy": 0.21341208666563033,
      "num_tokens": 88851016.0,
      "step": 42810
    },
    {
      "entropy": 5.461884689331055,
      "epoch": 3.8837989840348333,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003559800624078111,
      "loss": 5.0437,
      "mean_token_accuracy": 0.21190334260463714,
      "num_tokens": 88859795.0,
      "step": 42815
    },
    {
      "entropy": 5.436719608306885,
      "epoch": 3.884252539912917,
      "grad_norm": 1.0,
      "learning_rate": 0.0003559498734679949,
      "loss": 5.0567,
      "mean_token_accuracy": 0.2089402586221695,
      "num_tokens": 88869595.0,
      "step": 42820
    },
    {
      "entropy": 5.564501667022705,
      "epoch": 3.8847060957910013,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003559196828541148,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19894739240407944,
      "num_tokens": 88880857.0,
      "step": 42825
    },
    {
      "entropy": 5.465813112258911,
      "epoch": 3.8851596516690856,
      "grad_norm": 1.015625,
      "learning_rate": 0.000355889490566795,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21926327496767045,
      "num_tokens": 88890687.0,
      "step": 42830
    },
    {
      "entropy": 5.471407747268676,
      "epoch": 3.88561320754717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035585929660665984,
      "loss": 4.9903,
      "mean_token_accuracy": 0.21126246601343154,
      "num_tokens": 88899600.0,
      "step": 42835
    },
    {
      "entropy": 5.491221237182617,
      "epoch": 3.886066763425254,
      "grad_norm": 1.25,
      "learning_rate": 0.0003558291009743339,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20912323892116547,
      "num_tokens": 88909521.0,
      "step": 42840
    },
    {
      "entropy": 5.455891752243042,
      "epoch": 3.8865203193033384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035579890367044153,
      "loss": 5.0805,
      "mean_token_accuracy": 0.20357958674430848,
      "num_tokens": 88920061.0,
      "step": 42845
    },
    {
      "entropy": 5.393172740936279,
      "epoch": 3.8869738751814222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003557687046956073,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2109910950064659,
      "num_tokens": 88929504.0,
      "step": 42850
    },
    {
      "entropy": 5.577386283874512,
      "epoch": 3.8874274310595065,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00035573850405045554,
      "loss": 5.1545,
      "mean_token_accuracy": 0.19847708344459533,
      "num_tokens": 88940653.0,
      "step": 42855
    },
    {
      "entropy": 5.576539325714111,
      "epoch": 3.8878809869375908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003557083017356111,
      "loss": 5.0991,
      "mean_token_accuracy": 0.20643659830093383,
      "num_tokens": 88952133.0,
      "step": 42860
    },
    {
      "entropy": 5.42908239364624,
      "epoch": 3.888334542815675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003556780977516983,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2095394641160965,
      "num_tokens": 88963190.0,
      "step": 42865
    },
    {
      "entropy": 5.36001238822937,
      "epoch": 3.888788098693759,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035564789209934183,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21093632131814957,
      "num_tokens": 88973034.0,
      "step": 42870
    },
    {
      "entropy": 5.464149141311646,
      "epoch": 3.889241654571843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035561768477916655,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21824537962675095,
      "num_tokens": 88982809.0,
      "step": 42875
    },
    {
      "entropy": 5.47816047668457,
      "epoch": 3.8896952104499274,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000355587475791797,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20645194500684738,
      "num_tokens": 88993563.0,
      "step": 42880
    },
    {
      "entropy": 5.3665704250335695,
      "epoch": 3.8901487663280117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035555726513785785,
      "loss": 4.9301,
      "mean_token_accuracy": 0.214919114112854,
      "num_tokens": 89002408.0,
      "step": 42885
    },
    {
      "entropy": 5.417772245407105,
      "epoch": 3.890602322206096,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003555270528179739,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2148737907409668,
      "num_tokens": 89013060.0,
      "step": 42890
    },
    {
      "entropy": 5.5283989906311035,
      "epoch": 3.89105587808418,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035549683883277,
      "loss": 5.0523,
      "mean_token_accuracy": 0.20998213738203048,
      "num_tokens": 89023406.0,
      "step": 42895
    },
    {
      "entropy": 5.477376222610474,
      "epoch": 3.891509433962264,
      "grad_norm": 1.03125,
      "learning_rate": 0.000355466623182871,
      "loss": 5.0689,
      "mean_token_accuracy": 0.2103709101676941,
      "num_tokens": 89033800.0,
      "step": 42900
    },
    {
      "entropy": 5.542953681945801,
      "epoch": 3.8919629898403483,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003554364058689017,
      "loss": 5.0523,
      "mean_token_accuracy": 0.2077162057161331,
      "num_tokens": 89044601.0,
      "step": 42905
    },
    {
      "entropy": 5.580896186828613,
      "epoch": 3.8924165457184325,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000355406186891487,
      "loss": 5.0953,
      "mean_token_accuracy": 0.201276533305645,
      "num_tokens": 89056138.0,
      "step": 42910
    },
    {
      "entropy": 5.49587984085083,
      "epoch": 3.892870101596517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003553759662512519,
      "loss": 5.0057,
      "mean_token_accuracy": 0.21239876449108125,
      "num_tokens": 89065776.0,
      "step": 42915
    },
    {
      "entropy": 5.583881568908692,
      "epoch": 3.8933236574746006,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00035534574394882135,
      "loss": 5.1322,
      "mean_token_accuracy": 0.19506710916757583,
      "num_tokens": 89076926.0,
      "step": 42920
    },
    {
      "entropy": 5.494768524169922,
      "epoch": 3.893777213352685,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003553155199848203,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20089084506034852,
      "num_tokens": 89088036.0,
      "step": 42925
    },
    {
      "entropy": 5.46954493522644,
      "epoch": 3.894230769230769,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003552852943598739,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20620889216661453,
      "num_tokens": 89098606.0,
      "step": 42930
    },
    {
      "entropy": 5.462654399871826,
      "epoch": 3.8946843251088534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000355255067074607,
      "loss": 4.9685,
      "mean_token_accuracy": 0.21530644297599794,
      "num_tokens": 89107733.0,
      "step": 42935
    },
    {
      "entropy": 5.449081897735596,
      "epoch": 3.8951378809869377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035522483812964496,
      "loss": 4.9888,
      "mean_token_accuracy": 0.21360151320695878,
      "num_tokens": 89119094.0,
      "step": 42940
    },
    {
      "entropy": 5.448340272903442,
      "epoch": 3.895591436865022,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035519460752561285,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21544501483440398,
      "num_tokens": 89129254.0,
      "step": 42945
    },
    {
      "entropy": 5.447639894485474,
      "epoch": 3.8960449927431062,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003551643752631358,
      "loss": 4.9503,
      "mean_token_accuracy": 0.22040380984544755,
      "num_tokens": 89139609.0,
      "step": 42950
    },
    {
      "entropy": 5.463426828384399,
      "epoch": 3.89649854862119,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003551341413428391,
      "loss": 4.9804,
      "mean_token_accuracy": 0.2053520828485489,
      "num_tokens": 89149504.0,
      "step": 42955
    },
    {
      "entropy": 5.4897667407989506,
      "epoch": 3.8969521044992743,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003551039057653478,
      "loss": 5.0103,
      "mean_token_accuracy": 0.2126900315284729,
      "num_tokens": 89160036.0,
      "step": 42960
    },
    {
      "entropy": 5.469531917572022,
      "epoch": 3.8974056603773586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003550736685312874,
      "loss": 5.0602,
      "mean_token_accuracy": 0.19895631968975067,
      "num_tokens": 89170960.0,
      "step": 42965
    },
    {
      "entropy": 5.451873302459717,
      "epoch": 3.8978592162554424,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003550434296412831,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21491985470056535,
      "num_tokens": 89179630.0,
      "step": 42970
    },
    {
      "entropy": 5.4809221744537355,
      "epoch": 3.8983127721335267,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003550131890959603,
      "loss": 5.0115,
      "mean_token_accuracy": 0.2133575886487961,
      "num_tokens": 89188880.0,
      "step": 42975
    },
    {
      "entropy": 5.493407201766968,
      "epoch": 3.898766328011611,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035498294689594435,
      "loss": 5.0536,
      "mean_token_accuracy": 0.2080459862947464,
      "num_tokens": 89199585.0,
      "step": 42980
    },
    {
      "entropy": 5.450575256347657,
      "epoch": 3.899219883889695,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035495270304186074,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20387634187936782,
      "num_tokens": 89209812.0,
      "step": 42985
    },
    {
      "entropy": 5.474983072280883,
      "epoch": 3.8996734397677795,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003549224575343347,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21417993307113647,
      "num_tokens": 89220244.0,
      "step": 42990
    },
    {
      "entropy": 5.452927112579346,
      "epoch": 3.9001269956458637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035489221037399203,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20481956005096436,
      "num_tokens": 89230618.0,
      "step": 42995
    },
    {
      "entropy": 5.484771966934204,
      "epoch": 3.900580551523948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035486196156145805,
      "loss": 4.9908,
      "mean_token_accuracy": 0.2111361339688301,
      "num_tokens": 89242041.0,
      "step": 43000
    },
    {
      "entropy": 5.50086727142334,
      "epoch": 3.901034107402032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003548317110973583,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20701206773519515,
      "num_tokens": 89252817.0,
      "step": 43005
    },
    {
      "entropy": 5.515658521652222,
      "epoch": 3.901487663280116,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003548014589823185,
      "loss": 5.049,
      "mean_token_accuracy": 0.2134649708867073,
      "num_tokens": 89263702.0,
      "step": 43010
    },
    {
      "entropy": 5.447415590286255,
      "epoch": 3.9019412191582004,
      "grad_norm": 1.0,
      "learning_rate": 0.00035477120521696417,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21718219220638274,
      "num_tokens": 89273822.0,
      "step": 43015
    },
    {
      "entropy": 5.510671234130859,
      "epoch": 3.9023947750362846,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003547409498019209,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19736672341823577,
      "num_tokens": 89283973.0,
      "step": 43020
    },
    {
      "entropy": 5.512408113479614,
      "epoch": 3.9028483309143684,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035471069273781454,
      "loss": 5.0888,
      "mean_token_accuracy": 0.20973378717899321,
      "num_tokens": 89294059.0,
      "step": 43025
    },
    {
      "entropy": 5.463537216186523,
      "epoch": 3.9033018867924527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003546804340252707,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20966932773590088,
      "num_tokens": 89304049.0,
      "step": 43030
    },
    {
      "entropy": 5.412484121322632,
      "epoch": 3.903755442670537,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035465017366491527,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21894802451133727,
      "num_tokens": 89314468.0,
      "step": 43035
    },
    {
      "entropy": 5.415191555023194,
      "epoch": 3.9042089985486212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035461991165737386,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21457152664661408,
      "num_tokens": 89324580.0,
      "step": 43040
    },
    {
      "entropy": 5.523280954360962,
      "epoch": 3.9046625544267055,
      "grad_norm": 1.0,
      "learning_rate": 0.00035458964800327245,
      "loss": 5.0833,
      "mean_token_accuracy": 0.20026302933692933,
      "num_tokens": 89334976.0,
      "step": 43045
    },
    {
      "entropy": 5.500336027145385,
      "epoch": 3.90511611030479,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035455938270323684,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20787518620491027,
      "num_tokens": 89345210.0,
      "step": 43050
    },
    {
      "entropy": 5.429637145996094,
      "epoch": 3.9055696661828736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003545291157578928,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21737087666988372,
      "num_tokens": 89354623.0,
      "step": 43055
    },
    {
      "entropy": 5.457424116134644,
      "epoch": 3.906023222060958,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035449884716786653,
      "loss": 4.9856,
      "mean_token_accuracy": 0.2093849465250969,
      "num_tokens": 89364684.0,
      "step": 43060
    },
    {
      "entropy": 5.458658075332641,
      "epoch": 3.906476777939042,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00035446857693378377,
      "loss": 5.0406,
      "mean_token_accuracy": 0.20520026981830597,
      "num_tokens": 89375415.0,
      "step": 43065
    },
    {
      "entropy": 5.503015184402466,
      "epoch": 3.9069303338171264,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003544383050562706,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20869515389204024,
      "num_tokens": 89385791.0,
      "step": 43070
    },
    {
      "entropy": 5.5374244213104244,
      "epoch": 3.90738388969521,
      "grad_norm": 1.03125,
      "learning_rate": 0.000354408031535953,
      "loss": 5.1384,
      "mean_token_accuracy": 0.2040468782186508,
      "num_tokens": 89395345.0,
      "step": 43075
    },
    {
      "entropy": 5.474262619018555,
      "epoch": 3.9078374455732945,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035437775637345715,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21497057974338532,
      "num_tokens": 89405413.0,
      "step": 43080
    },
    {
      "entropy": 5.49092059135437,
      "epoch": 3.9082910014513788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035434747956940906,
      "loss": 5.0017,
      "mean_token_accuracy": 0.21679402887821198,
      "num_tokens": 89415168.0,
      "step": 43085
    },
    {
      "entropy": 5.4301306247711185,
      "epoch": 3.908744557329463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035431720112443475,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20890072584152222,
      "num_tokens": 89426438.0,
      "step": 43090
    },
    {
      "entropy": 5.4440103530883786,
      "epoch": 3.9091981132075473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035428692103916067,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20464864373207092,
      "num_tokens": 89437454.0,
      "step": 43095
    },
    {
      "entropy": 5.543535280227661,
      "epoch": 3.9096516690856316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003542566393142128,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21131522655487062,
      "num_tokens": 89448082.0,
      "step": 43100
    },
    {
      "entropy": 5.392830038070679,
      "epoch": 3.9101052249637154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035422635595021745,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21421265304088594,
      "num_tokens": 89458227.0,
      "step": 43105
    },
    {
      "entropy": 5.419227504730225,
      "epoch": 3.9105587808417996,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00035419607094780084,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20864237844944,
      "num_tokens": 89468374.0,
      "step": 43110
    },
    {
      "entropy": 5.486902284622192,
      "epoch": 3.911012336719884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035416578430758934,
      "loss": 5.0602,
      "mean_token_accuracy": 0.21064115911722184,
      "num_tokens": 89477728.0,
      "step": 43115
    },
    {
      "entropy": 5.515620326995849,
      "epoch": 3.911465892597968,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00035413549603020915,
      "loss": 5.0322,
      "mean_token_accuracy": 0.2076970249414444,
      "num_tokens": 89487107.0,
      "step": 43120
    },
    {
      "entropy": 5.511048603057861,
      "epoch": 3.911919448476052,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035410520611628676,
      "loss": 5.05,
      "mean_token_accuracy": 0.21212968677282334,
      "num_tokens": 89497075.0,
      "step": 43125
    },
    {
      "entropy": 5.406746196746826,
      "epoch": 3.9123730043541363,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003540749145664486,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21049065589904786,
      "num_tokens": 89508068.0,
      "step": 43130
    },
    {
      "entropy": 5.493825626373291,
      "epoch": 3.9128265602322205,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003540446213813211,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2211065486073494,
      "num_tokens": 89518371.0,
      "step": 43135
    },
    {
      "entropy": 5.481895780563354,
      "epoch": 3.913280116110305,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035401432656153057,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21076659560203553,
      "num_tokens": 89529024.0,
      "step": 43140
    },
    {
      "entropy": 5.390733814239502,
      "epoch": 3.913733671988389,
      "grad_norm": 1.0,
      "learning_rate": 0.00035398403010770377,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21568010598421097,
      "num_tokens": 89540331.0,
      "step": 43145
    },
    {
      "entropy": 5.3604405403137205,
      "epoch": 3.9141872278664733,
      "grad_norm": 1.015625,
      "learning_rate": 0.000353953732020467,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21965165585279464,
      "num_tokens": 89550385.0,
      "step": 43150
    },
    {
      "entropy": 5.5112464427948,
      "epoch": 3.9146407837445576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035392343230044687,
      "loss": 5.1018,
      "mean_token_accuracy": 0.19814615249633788,
      "num_tokens": 89561480.0,
      "step": 43155
    },
    {
      "entropy": 5.486927270889282,
      "epoch": 3.9150943396226414,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035389313094827017,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20461529940366746,
      "num_tokens": 89571897.0,
      "step": 43160
    },
    {
      "entropy": 5.511381721496582,
      "epoch": 3.9155478955007257,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035386282796456343,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2111704558134079,
      "num_tokens": 89582019.0,
      "step": 43165
    },
    {
      "entropy": 5.484508371353149,
      "epoch": 3.91600145137881,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003538325233499532,
      "loss": 4.9991,
      "mean_token_accuracy": 0.2128277912735939,
      "num_tokens": 89591806.0,
      "step": 43170
    },
    {
      "entropy": 5.435688257217407,
      "epoch": 3.9164550072568938,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035380221710506636,
      "loss": 5.0102,
      "mean_token_accuracy": 0.2070450410246849,
      "num_tokens": 89601845.0,
      "step": 43175
    },
    {
      "entropy": 5.479626178741455,
      "epoch": 3.916908563134978,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00035377190923052964,
      "loss": 5.0788,
      "mean_token_accuracy": 0.20625646263360978,
      "num_tokens": 89612331.0,
      "step": 43180
    },
    {
      "entropy": 5.543519783020019,
      "epoch": 3.9173621190130623,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035374159972696966,
      "loss": 5.0764,
      "mean_token_accuracy": 0.20324315279722213,
      "num_tokens": 89622631.0,
      "step": 43185
    },
    {
      "entropy": 5.515856170654297,
      "epoch": 3.9178156748911466,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003537112885950133,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20301297158002854,
      "num_tokens": 89633065.0,
      "step": 43190
    },
    {
      "entropy": 5.505182695388794,
      "epoch": 3.918269230769231,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035368097583528756,
      "loss": 5.0596,
      "mean_token_accuracy": 0.2065563678741455,
      "num_tokens": 89643451.0,
      "step": 43195
    },
    {
      "entropy": 5.569129085540771,
      "epoch": 3.918722786647315,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003536506614484192,
      "loss": 5.0772,
      "mean_token_accuracy": 0.20526812672615052,
      "num_tokens": 89654786.0,
      "step": 43200
    },
    {
      "entropy": 5.520526170730591,
      "epoch": 3.9191763425253994,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003536203454350351,
      "loss": 5.0574,
      "mean_token_accuracy": 0.2092939719557762,
      "num_tokens": 89665459.0,
      "step": 43205
    },
    {
      "entropy": 5.436714553833008,
      "epoch": 3.919629898403483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003535900277957621,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2146658033132553,
      "num_tokens": 89675149.0,
      "step": 43210
    },
    {
      "entropy": 5.468516540527344,
      "epoch": 3.9200834542815675,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035355970853122743,
      "loss": 5.0412,
      "mean_token_accuracy": 0.21347324103116988,
      "num_tokens": 89684994.0,
      "step": 43215
    },
    {
      "entropy": 5.550519323348999,
      "epoch": 3.9205370101596517,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035352938764205793,
      "loss": 5.148,
      "mean_token_accuracy": 0.19703109115362166,
      "num_tokens": 89694751.0,
      "step": 43220
    },
    {
      "entropy": 5.52348952293396,
      "epoch": 3.920990566037736,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003534990651288807,
      "loss": 4.976,
      "mean_token_accuracy": 0.20863630175590514,
      "num_tokens": 89704976.0,
      "step": 43225
    },
    {
      "entropy": 5.514065885543824,
      "epoch": 3.92144412191582,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035346874099232274,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19722842425107956,
      "num_tokens": 89715863.0,
      "step": 43230
    },
    {
      "entropy": 5.461881685256958,
      "epoch": 3.921897677793904,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035343841523301125,
      "loss": 5.0657,
      "mean_token_accuracy": 0.20783078968524932,
      "num_tokens": 89725856.0,
      "step": 43235
    },
    {
      "entropy": 5.494053745269776,
      "epoch": 3.9223512336719883,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003534080878515733,
      "loss": 5.0742,
      "mean_token_accuracy": 0.20173437148332596,
      "num_tokens": 89736345.0,
      "step": 43240
    },
    {
      "entropy": 5.5620862483978275,
      "epoch": 3.9228047895500726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003533777588486362,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20940585434436798,
      "num_tokens": 89746960.0,
      "step": 43245
    },
    {
      "entropy": 5.454016590118409,
      "epoch": 3.923258345428157,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000353347428224827,
      "loss": 5.046,
      "mean_token_accuracy": 0.2096942812204361,
      "num_tokens": 89757373.0,
      "step": 43250
    },
    {
      "entropy": 5.510996532440186,
      "epoch": 3.923711901306241,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035331709598077304,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20489539057016373,
      "num_tokens": 89767895.0,
      "step": 43255
    },
    {
      "entropy": 5.552409839630127,
      "epoch": 3.924165457184325,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003532867621171016,
      "loss": 5.0072,
      "mean_token_accuracy": 0.21059472113847733,
      "num_tokens": 89777512.0,
      "step": 43260
    },
    {
      "entropy": 5.444802522659302,
      "epoch": 3.9246190130624092,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035325642663443994,
      "loss": 5.0451,
      "mean_token_accuracy": 0.21010637283325195,
      "num_tokens": 89787308.0,
      "step": 43265
    },
    {
      "entropy": 5.4974956035614015,
      "epoch": 3.9250725689404935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003532260895334155,
      "loss": 5.0564,
      "mean_token_accuracy": 0.20966505408287048,
      "num_tokens": 89798384.0,
      "step": 43270
    },
    {
      "entropy": 5.53019905090332,
      "epoch": 3.9255261248185778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035319575081465556,
      "loss": 5.0857,
      "mean_token_accuracy": 0.20913304537534713,
      "num_tokens": 89808757.0,
      "step": 43275
    },
    {
      "entropy": 5.53930287361145,
      "epoch": 3.9259796806966616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035316541047878753,
      "loss": 5.0451,
      "mean_token_accuracy": 0.20777703672647477,
      "num_tokens": 89818925.0,
      "step": 43280
    },
    {
      "entropy": 5.507578134536743,
      "epoch": 3.926433236574746,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003531350685264389,
      "loss": 5.094,
      "mean_token_accuracy": 0.21184618026018143,
      "num_tokens": 89828647.0,
      "step": 43285
    },
    {
      "entropy": 5.5690497875213625,
      "epoch": 3.92688679245283,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00035310472495823713,
      "loss": 5.127,
      "mean_token_accuracy": 0.20004664957523347,
      "num_tokens": 89838784.0,
      "step": 43290
    },
    {
      "entropy": 5.477791023254395,
      "epoch": 3.9273403483309144,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003530743797748098,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2012937471270561,
      "num_tokens": 89848464.0,
      "step": 43295
    },
    {
      "entropy": 5.468021488189697,
      "epoch": 3.9277939042089987,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003530440329767844,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1985919803380966,
      "num_tokens": 89859594.0,
      "step": 43300
    },
    {
      "entropy": 5.525835704803467,
      "epoch": 3.928247460087083,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003530136845647885,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21009097397327423,
      "num_tokens": 89869869.0,
      "step": 43305
    },
    {
      "entropy": 5.573779106140137,
      "epoch": 3.9287010159651667,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035298333453944976,
      "loss": 5.0767,
      "mean_token_accuracy": 0.2117905616760254,
      "num_tokens": 89879604.0,
      "step": 43310
    },
    {
      "entropy": 5.323205852508545,
      "epoch": 3.929154571843251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003529529829013957,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2147512450814247,
      "num_tokens": 89889200.0,
      "step": 43315
    },
    {
      "entropy": 5.463785839080811,
      "epoch": 3.9296081277213353,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003529226296512541,
      "loss": 5.0306,
      "mean_token_accuracy": 0.21002936959266663,
      "num_tokens": 89898525.0,
      "step": 43320
    },
    {
      "entropy": 5.456488132476807,
      "epoch": 3.9300616835994195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035289227478965267,
      "loss": 5.0664,
      "mean_token_accuracy": 0.20470192432403564,
      "num_tokens": 89909053.0,
      "step": 43325
    },
    {
      "entropy": 5.427297782897949,
      "epoch": 3.9305152394775034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003528619183172192,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21888616681098938,
      "num_tokens": 89918375.0,
      "step": 43330
    },
    {
      "entropy": 5.556090688705444,
      "epoch": 3.9309687953555876,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003528315602345814,
      "loss": 5.0948,
      "mean_token_accuracy": 0.21223596781492232,
      "num_tokens": 89928252.0,
      "step": 43335
    },
    {
      "entropy": 5.442232084274292,
      "epoch": 3.931422351233672,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003528012005423671,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20369129925966262,
      "num_tokens": 89938613.0,
      "step": 43340
    },
    {
      "entropy": 5.522843885421753,
      "epoch": 3.931875907111756,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035277083924120406,
      "loss": 5.0498,
      "mean_token_accuracy": 0.2084174782037735,
      "num_tokens": 89948781.0,
      "step": 43345
    },
    {
      "entropy": 5.503701257705688,
      "epoch": 3.9323294629898404,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035274047633172037,
      "loss": 5.0536,
      "mean_token_accuracy": 0.2036526992917061,
      "num_tokens": 89959928.0,
      "step": 43350
    },
    {
      "entropy": 5.527623271942138,
      "epoch": 3.9327830188679247,
      "grad_norm": 1.0,
      "learning_rate": 0.0003527101118145437,
      "loss": 5.0935,
      "mean_token_accuracy": 0.2042301520705223,
      "num_tokens": 89970756.0,
      "step": 43355
    },
    {
      "entropy": 5.51734938621521,
      "epoch": 3.933236574746009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003526797456903022,
      "loss": 5.0583,
      "mean_token_accuracy": 0.20842934399843216,
      "num_tokens": 89981332.0,
      "step": 43360
    },
    {
      "entropy": 5.464041328430175,
      "epoch": 3.933690130624093,
      "grad_norm": 0.984375,
      "learning_rate": 0.00035264937795962374,
      "loss": 5.0106,
      "mean_token_accuracy": 0.2125599965453148,
      "num_tokens": 89991427.0,
      "step": 43365
    },
    {
      "entropy": 5.523365879058838,
      "epoch": 3.934143686502177,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003526190086231364,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21360450834035874,
      "num_tokens": 90001599.0,
      "step": 43370
    },
    {
      "entropy": 5.529808759689331,
      "epoch": 3.9345972423802613,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035258863768146816,
      "loss": 5.028,
      "mean_token_accuracy": 0.21141710579395295,
      "num_tokens": 90011932.0,
      "step": 43375
    },
    {
      "entropy": 5.4525714874267575,
      "epoch": 3.9350507982583456,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003525582651352471,
      "loss": 5.1169,
      "mean_token_accuracy": 0.20715624392032622,
      "num_tokens": 90022123.0,
      "step": 43380
    },
    {
      "entropy": 5.3843944549560545,
      "epoch": 3.9355043541364294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003525278909851013,
      "loss": 4.9693,
      "mean_token_accuracy": 0.2190212145447731,
      "num_tokens": 90033024.0,
      "step": 43385
    },
    {
      "entropy": 5.517726087570191,
      "epoch": 3.9359579100145137,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000352497515231659,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21330399364233016,
      "num_tokens": 90043742.0,
      "step": 43390
    },
    {
      "entropy": 5.476689577102661,
      "epoch": 3.936411465892598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003524671378755484,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20881394892930985,
      "num_tokens": 90054508.0,
      "step": 43395
    },
    {
      "entropy": 5.448577260971069,
      "epoch": 3.936865021770682,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003524367589173976,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21075050085783004,
      "num_tokens": 90065058.0,
      "step": 43400
    },
    {
      "entropy": 5.4751801013946535,
      "epoch": 3.9373185776487665,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035240637835783485,
      "loss": 5.0557,
      "mean_token_accuracy": 0.21075084656476975,
      "num_tokens": 90075813.0,
      "step": 43405
    },
    {
      "entropy": 5.547541618347168,
      "epoch": 3.9377721335268507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035237599619748853,
      "loss": 5.1128,
      "mean_token_accuracy": 0.20018366128206252,
      "num_tokens": 90087064.0,
      "step": 43410
    },
    {
      "entropy": 5.471080303192139,
      "epoch": 3.9382256894049346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003523456124369868,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2139293819665909,
      "num_tokens": 90096668.0,
      "step": 43415
    },
    {
      "entropy": 5.413409042358398,
      "epoch": 3.938679245283019,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003523152270769582,
      "loss": 4.8928,
      "mean_token_accuracy": 0.21905346512794494,
      "num_tokens": 90105569.0,
      "step": 43420
    },
    {
      "entropy": 5.480507850646973,
      "epoch": 3.939132801161103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000352284840118031,
      "loss": 5.0663,
      "mean_token_accuracy": 0.20490254312753678,
      "num_tokens": 90116040.0,
      "step": 43425
    },
    {
      "entropy": 5.498834609985352,
      "epoch": 3.9395863570391874,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003522544515608335,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20774238407611847,
      "num_tokens": 90125929.0,
      "step": 43430
    },
    {
      "entropy": 5.5972686290740965,
      "epoch": 3.940039912917271,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003522240614059944,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2057241603732109,
      "num_tokens": 90136251.0,
      "step": 43435
    },
    {
      "entropy": 5.484125709533691,
      "epoch": 3.9404934687953554,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000352193669654142,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2134733498096466,
      "num_tokens": 90146480.0,
      "step": 43440
    },
    {
      "entropy": 5.519536304473877,
      "epoch": 3.9409470246734397,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035216327630590476,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20912301540374756,
      "num_tokens": 90157192.0,
      "step": 43445
    },
    {
      "entropy": 5.503592252731323,
      "epoch": 3.941400580551524,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035213288136191136,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20731087476015092,
      "num_tokens": 90167120.0,
      "step": 43450
    },
    {
      "entropy": 5.464799928665161,
      "epoch": 3.9418541364296082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035210248482279034,
      "loss": 5.0571,
      "mean_token_accuracy": 0.21209718734025956,
      "num_tokens": 90176276.0,
      "step": 43455
    },
    {
      "entropy": 5.472534990310669,
      "epoch": 3.9423076923076925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035207208668917026,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21044944673776628,
      "num_tokens": 90187150.0,
      "step": 43460
    },
    {
      "entropy": 5.437232494354248,
      "epoch": 3.9427612481857763,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035204168696167975,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20432120561599731,
      "num_tokens": 90198112.0,
      "step": 43465
    },
    {
      "entropy": 5.48224105834961,
      "epoch": 3.9432148040638606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035201128564094753,
      "loss": 5.0837,
      "mean_token_accuracy": 0.20952948182821274,
      "num_tokens": 90208940.0,
      "step": 43470
    },
    {
      "entropy": 5.521867465972901,
      "epoch": 3.943668359941945,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003519808827276023,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2103973150253296,
      "num_tokens": 90220073.0,
      "step": 43475
    },
    {
      "entropy": 5.484498167037964,
      "epoch": 3.944121915820029,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035195047822227283,
      "loss": 5.0452,
      "mean_token_accuracy": 0.2086349442601204,
      "num_tokens": 90230059.0,
      "step": 43480
    },
    {
      "entropy": 5.413075971603393,
      "epoch": 3.944575471698113,
      "grad_norm": 1.0,
      "learning_rate": 0.00035192007212558787,
      "loss": 4.9822,
      "mean_token_accuracy": 0.2116161599755287,
      "num_tokens": 90240486.0,
      "step": 43485
    },
    {
      "entropy": 5.5598071098327635,
      "epoch": 3.9450290275761972,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003518896644381762,
      "loss": 5.084,
      "mean_token_accuracy": 0.20590721666812897,
      "num_tokens": 90250014.0,
      "step": 43490
    },
    {
      "entropy": 5.525039100646973,
      "epoch": 3.9454825834542815,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003518592551606666,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20258343815803528,
      "num_tokens": 90261040.0,
      "step": 43495
    },
    {
      "entropy": 5.536734867095947,
      "epoch": 3.9459361393323658,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003518288442936881,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2105329915881157,
      "num_tokens": 90271243.0,
      "step": 43500
    },
    {
      "entropy": 5.439156436920166,
      "epoch": 3.94638969521045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003517984318378695,
      "loss": 5.0376,
      "mean_token_accuracy": 0.20333140343427658,
      "num_tokens": 90281176.0,
      "step": 43505
    },
    {
      "entropy": 5.628147888183594,
      "epoch": 3.9468432510885343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003517680177938398,
      "loss": 5.1191,
      "mean_token_accuracy": 0.205084690451622,
      "num_tokens": 90292465.0,
      "step": 43510
    },
    {
      "entropy": 5.501614379882812,
      "epoch": 3.9472968069666186,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003517376021622279,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21452865600585938,
      "num_tokens": 90303482.0,
      "step": 43515
    },
    {
      "entropy": 5.43652925491333,
      "epoch": 3.9477503628447024,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003517071849436629,
      "loss": 5.054,
      "mean_token_accuracy": 0.20892066061496734,
      "num_tokens": 90313696.0,
      "step": 43520
    },
    {
      "entropy": 5.5258441925048825,
      "epoch": 3.9482039187227866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035167676613877366,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20407017320394516,
      "num_tokens": 90324564.0,
      "step": 43525
    },
    {
      "entropy": 5.456768989562988,
      "epoch": 3.948657474600871,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003516463457481894,
      "loss": 4.904,
      "mean_token_accuracy": 0.21870799064636232,
      "num_tokens": 90334440.0,
      "step": 43530
    },
    {
      "entropy": 5.474437570571899,
      "epoch": 3.9491110304789547,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003516159237725392,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20840011686086654,
      "num_tokens": 90344029.0,
      "step": 43535
    },
    {
      "entropy": 5.5259437084198,
      "epoch": 3.949564586357039,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003515855002124521,
      "loss": 5.0659,
      "mean_token_accuracy": 0.2032058760523796,
      "num_tokens": 90354628.0,
      "step": 43540
    },
    {
      "entropy": 5.495509958267212,
      "epoch": 3.9500181422351233,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003515550750685573,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21361101269721985,
      "num_tokens": 90365091.0,
      "step": 43545
    },
    {
      "entropy": 5.442817974090576,
      "epoch": 3.9504716981132075,
      "grad_norm": 1.0,
      "learning_rate": 0.0003515246483414842,
      "loss": 5.034,
      "mean_token_accuracy": 0.20843005925416946,
      "num_tokens": 90374197.0,
      "step": 43550
    },
    {
      "entropy": 5.508985328674316,
      "epoch": 3.950925253991292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035149422003186176,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2103315144777298,
      "num_tokens": 90384686.0,
      "step": 43555
    },
    {
      "entropy": 5.562664270401001,
      "epoch": 3.951378809869376,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035146379014031943,
      "loss": 5.1008,
      "mean_token_accuracy": 0.20516750067472458,
      "num_tokens": 90396329.0,
      "step": 43560
    },
    {
      "entropy": 5.586495351791382,
      "epoch": 3.9518323657474603,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003514333586674864,
      "loss": 5.1332,
      "mean_token_accuracy": 0.20229458063840866,
      "num_tokens": 90408957.0,
      "step": 43565
    },
    {
      "entropy": 5.443402719497681,
      "epoch": 3.952285921625544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000351402925613992,
      "loss": 4.9706,
      "mean_token_accuracy": 0.2077053427696228,
      "num_tokens": 90420194.0,
      "step": 43570
    },
    {
      "entropy": 5.443138170242309,
      "epoch": 3.9527394775036284,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003513724909804656,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21365591138601303,
      "num_tokens": 90430881.0,
      "step": 43575
    },
    {
      "entropy": 5.452688694000244,
      "epoch": 3.9531930333817127,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035134205476753674,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20632417500019073,
      "num_tokens": 90440713.0,
      "step": 43580
    },
    {
      "entropy": 5.4994447231292725,
      "epoch": 3.953646589259797,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003513116169758347,
      "loss": 5.0818,
      "mean_token_accuracy": 0.20743847042322158,
      "num_tokens": 90451274.0,
      "step": 43585
    },
    {
      "entropy": 5.565321016311645,
      "epoch": 3.9541001451378808,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003512811776059889,
      "loss": 5.1237,
      "mean_token_accuracy": 0.2038185715675354,
      "num_tokens": 90461503.0,
      "step": 43590
    },
    {
      "entropy": 5.493304777145386,
      "epoch": 3.954553701015965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000351250736658629,
      "loss": 4.9699,
      "mean_token_accuracy": 0.21548430025577545,
      "num_tokens": 90471178.0,
      "step": 43595
    },
    {
      "entropy": 5.484933996200562,
      "epoch": 3.9550072568940493,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003512202941343844,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20872606039047242,
      "num_tokens": 90481573.0,
      "step": 43600
    },
    {
      "entropy": 5.481135606765747,
      "epoch": 3.9554608127721336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003511898500338847,
      "loss": 5.0808,
      "mean_token_accuracy": 0.2009224146604538,
      "num_tokens": 90491853.0,
      "step": 43605
    },
    {
      "entropy": 5.511242485046386,
      "epoch": 3.955914368650218,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003511594043577594,
      "loss": 5.0679,
      "mean_token_accuracy": 0.2032613068819046,
      "num_tokens": 90502958.0,
      "step": 43610
    },
    {
      "entropy": 5.439742183685302,
      "epoch": 3.956367924528302,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035112895710663833,
      "loss": 5.004,
      "mean_token_accuracy": 0.20818913578987122,
      "num_tokens": 90512555.0,
      "step": 43615
    },
    {
      "entropy": 5.516923952102661,
      "epoch": 3.956821480406386,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035109850828115105,
      "loss": 5.0437,
      "mean_token_accuracy": 0.21366630792617797,
      "num_tokens": 90522722.0,
      "step": 43620
    },
    {
      "entropy": 5.487954187393188,
      "epoch": 3.95727503628447,
      "grad_norm": 1.0,
      "learning_rate": 0.00035106805788192707,
      "loss": 5.0542,
      "mean_token_accuracy": 0.2083281695842743,
      "num_tokens": 90533182.0,
      "step": 43625
    },
    {
      "entropy": 5.514346122741699,
      "epoch": 3.9577285921625545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003510376059095963,
      "loss": 5.067,
      "mean_token_accuracy": 0.20186058878898622,
      "num_tokens": 90543495.0,
      "step": 43630
    },
    {
      "entropy": 5.428491449356079,
      "epoch": 3.9581821480406387,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003510071523647886,
      "loss": 4.9466,
      "mean_token_accuracy": 0.21360282450914383,
      "num_tokens": 90553680.0,
      "step": 43635
    },
    {
      "entropy": 5.455829334259033,
      "epoch": 3.9586357039187225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003509766972481336,
      "loss": 4.9331,
      "mean_token_accuracy": 0.20980989784002305,
      "num_tokens": 90563513.0,
      "step": 43640
    },
    {
      "entropy": 5.573322010040283,
      "epoch": 3.959089259796807,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000350946240560261,
      "loss": 5.0796,
      "mean_token_accuracy": 0.20564778447151183,
      "num_tokens": 90573337.0,
      "step": 43645
    },
    {
      "entropy": 5.419825601577759,
      "epoch": 3.959542815674891,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003509157823018009,
      "loss": 4.9982,
      "mean_token_accuracy": 0.2131268411874771,
      "num_tokens": 90582979.0,
      "step": 43650
    },
    {
      "entropy": 5.573810291290283,
      "epoch": 3.9599963715529753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035088532247338306,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19808611869812012,
      "num_tokens": 90593552.0,
      "step": 43655
    },
    {
      "entropy": 5.540936517715454,
      "epoch": 3.9604499274310596,
      "grad_norm": 0.984375,
      "learning_rate": 0.00035085486107563733,
      "loss": 5.0476,
      "mean_token_accuracy": 0.2077673926949501,
      "num_tokens": 90603813.0,
      "step": 43660
    },
    {
      "entropy": 5.500109672546387,
      "epoch": 3.960903483309144,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035082439810919385,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2137461170554161,
      "num_tokens": 90613885.0,
      "step": 43665
    },
    {
      "entropy": 5.492814207077027,
      "epoch": 3.9613570391872277,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003507939335746824,
      "loss": 4.9319,
      "mean_token_accuracy": 0.20928327590227128,
      "num_tokens": 90624665.0,
      "step": 43670
    },
    {
      "entropy": 5.4749291896820065,
      "epoch": 3.961810595065312,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003507634674727332,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21143167167901994,
      "num_tokens": 90635496.0,
      "step": 43675
    },
    {
      "entropy": 5.494802331924438,
      "epoch": 3.9622641509433962,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003507329998039762,
      "loss": 4.998,
      "mean_token_accuracy": 0.20840945839881897,
      "num_tokens": 90645502.0,
      "step": 43680
    },
    {
      "entropy": 5.410250043869018,
      "epoch": 3.9627177068214805,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035070253056904133,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21818565875291823,
      "num_tokens": 90654724.0,
      "step": 43685
    },
    {
      "entropy": 5.391813945770264,
      "epoch": 3.9631712626995643,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003506720597685589,
      "loss": 4.9829,
      "mean_token_accuracy": 0.21180761456489564,
      "num_tokens": 90665280.0,
      "step": 43690
    },
    {
      "entropy": 5.467523431777954,
      "epoch": 3.9636248185776486,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035064158740315894,
      "loss": 5.0572,
      "mean_token_accuracy": 0.20712157636880874,
      "num_tokens": 90675804.0,
      "step": 43695
    },
    {
      "entropy": 5.525162696838379,
      "epoch": 3.964078374455733,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00035061111347347174,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21228004842996598,
      "num_tokens": 90686636.0,
      "step": 43700
    },
    {
      "entropy": 5.4618682861328125,
      "epoch": 3.964531930333817,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035058063798012746,
      "loss": 5.0142,
      "mean_token_accuracy": 0.2120109125971794,
      "num_tokens": 90697666.0,
      "step": 43705
    },
    {
      "entropy": 5.544947624206543,
      "epoch": 3.9649854862119014,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003505501609237563,
      "loss": 5.1281,
      "mean_token_accuracy": 0.20083957612514497,
      "num_tokens": 90708111.0,
      "step": 43710
    },
    {
      "entropy": 5.4638162612915036,
      "epoch": 3.9654390420899857,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035051968230498853,
      "loss": 4.968,
      "mean_token_accuracy": 0.2150312751531601,
      "num_tokens": 90718711.0,
      "step": 43715
    },
    {
      "entropy": 5.414331483840942,
      "epoch": 3.96589259796807,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035048920212445454,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21483734250068665,
      "num_tokens": 90729585.0,
      "step": 43720
    },
    {
      "entropy": 5.396825075149536,
      "epoch": 3.9663461538461537,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035045872038278467,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22064369022846222,
      "num_tokens": 90739654.0,
      "step": 43725
    },
    {
      "entropy": 5.470814323425293,
      "epoch": 3.966799709724238,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003504282370806091,
      "loss": 4.9641,
      "mean_token_accuracy": 0.2143012821674347,
      "num_tokens": 90749456.0,
      "step": 43730
    },
    {
      "entropy": 5.462963724136353,
      "epoch": 3.9672532656023223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035039775221855845,
      "loss": 4.9724,
      "mean_token_accuracy": 0.21365687102079392,
      "num_tokens": 90760051.0,
      "step": 43735
    },
    {
      "entropy": 5.418946123123169,
      "epoch": 3.9677068214804065,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000350367265797263,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21642662435770035,
      "num_tokens": 90770452.0,
      "step": 43740
    },
    {
      "entropy": 5.469805288314819,
      "epoch": 3.9681603773584904,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035033677781735336,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21215900033712387,
      "num_tokens": 90780150.0,
      "step": 43745
    },
    {
      "entropy": 5.4505539894104,
      "epoch": 3.9686139332365746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003503062882794599,
      "loss": 4.9493,
      "mean_token_accuracy": 0.22011968195438386,
      "num_tokens": 90790338.0,
      "step": 43750
    },
    {
      "entropy": 5.444595670700073,
      "epoch": 3.969067489114659,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035027579718421327,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20946687161922456,
      "num_tokens": 90800195.0,
      "step": 43755
    },
    {
      "entropy": 5.515007972717285,
      "epoch": 3.969521044992743,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003502453045322439,
      "loss": 5.0824,
      "mean_token_accuracy": 0.20368384718894958,
      "num_tokens": 90810872.0,
      "step": 43760
    },
    {
      "entropy": 5.496774768829345,
      "epoch": 3.9699746008708274,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035021481032418254,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21320506930351257,
      "num_tokens": 90821018.0,
      "step": 43765
    },
    {
      "entropy": 5.441741466522217,
      "epoch": 3.9704281567489117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003501843145606597,
      "loss": 5.0278,
      "mean_token_accuracy": 0.2071130245923996,
      "num_tokens": 90830274.0,
      "step": 43770
    },
    {
      "entropy": 5.426712894439698,
      "epoch": 3.9708817126269955,
      "grad_norm": 1.015625,
      "learning_rate": 0.000350153817242306,
      "loss": 4.934,
      "mean_token_accuracy": 0.2209562450647354,
      "num_tokens": 90839926.0,
      "step": 43775
    },
    {
      "entropy": 5.407787895202636,
      "epoch": 3.97133526850508,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003501233183697523,
      "loss": 4.9,
      "mean_token_accuracy": 0.21112582385540007,
      "num_tokens": 90850263.0,
      "step": 43780
    },
    {
      "entropy": 5.488572978973389,
      "epoch": 3.971788824383164,
      "grad_norm": 1.0,
      "learning_rate": 0.00035009281794362926,
      "loss": 5.0216,
      "mean_token_accuracy": 0.200329726934433,
      "num_tokens": 90861747.0,
      "step": 43785
    },
    {
      "entropy": 5.499306344985962,
      "epoch": 3.9722423802612483,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00035006231596456755,
      "loss": 5.0911,
      "mean_token_accuracy": 0.2009745329618454,
      "num_tokens": 90872186.0,
      "step": 43790
    },
    {
      "entropy": 5.504335260391235,
      "epoch": 3.972695936139332,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035003181243319797,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21384303271770477,
      "num_tokens": 90881088.0,
      "step": 43795
    },
    {
      "entropy": 5.384993648529052,
      "epoch": 3.9731494920174164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003500013073501514,
      "loss": 4.9512,
      "mean_token_accuracy": 0.20529115945100784,
      "num_tokens": 90890589.0,
      "step": 43800
    },
    {
      "entropy": 5.454790258407593,
      "epoch": 3.9736030478955007,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003499708007160588,
      "loss": 5.076,
      "mean_token_accuracy": 0.20995931476354598,
      "num_tokens": 90901692.0,
      "step": 43805
    },
    {
      "entropy": 5.541680192947387,
      "epoch": 3.974056603773585,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003499402925315509,
      "loss": 5.0861,
      "mean_token_accuracy": 0.20751251578330993,
      "num_tokens": 90911811.0,
      "step": 43810
    },
    {
      "entropy": 5.489838790893555,
      "epoch": 3.974510159651669,
      "grad_norm": 1.0,
      "learning_rate": 0.00034990978279725846,
      "loss": 4.9159,
      "mean_token_accuracy": 0.21587300598621367,
      "num_tokens": 90922323.0,
      "step": 43815
    },
    {
      "entropy": 5.560454511642456,
      "epoch": 3.9749637155297535,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034987927151381283,
      "loss": 5.1975,
      "mean_token_accuracy": 0.19855079054832458,
      "num_tokens": 90932834.0,
      "step": 43820
    },
    {
      "entropy": 5.498272895812988,
      "epoch": 3.9754172714078373,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034984875868184466,
      "loss": 5.0108,
      "mean_token_accuracy": 0.21072230041027068,
      "num_tokens": 90943616.0,
      "step": 43825
    },
    {
      "entropy": 5.524191570281983,
      "epoch": 3.9758708272859216,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034981824430198517,
      "loss": 5.0853,
      "mean_token_accuracy": 0.2036605030298233,
      "num_tokens": 90954468.0,
      "step": 43830
    },
    {
      "entropy": 5.491626739501953,
      "epoch": 3.976324383164006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034978772837486536,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21804403960704805,
      "num_tokens": 90965109.0,
      "step": 43835
    },
    {
      "entropy": 5.503322267532349,
      "epoch": 3.97677793904209,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034975721090111625,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20442949533462523,
      "num_tokens": 90974789.0,
      "step": 43840
    },
    {
      "entropy": 5.496401357650757,
      "epoch": 3.977231494920174,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003497266918813689,
      "loss": 5.0902,
      "mean_token_accuracy": 0.208639295399189,
      "num_tokens": 90985154.0,
      "step": 43845
    },
    {
      "entropy": 5.476354598999023,
      "epoch": 3.977685050798258,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003496961713162545,
      "loss": 5.0564,
      "mean_token_accuracy": 0.2057189702987671,
      "num_tokens": 90996007.0,
      "step": 43850
    },
    {
      "entropy": 5.508702659606934,
      "epoch": 3.9781386066763424,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003496656492064043,
      "loss": 5.0794,
      "mean_token_accuracy": 0.20181174725294113,
      "num_tokens": 91006253.0,
      "step": 43855
    },
    {
      "entropy": 5.569933938980102,
      "epoch": 3.9785921625544267,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003496351255524494,
      "loss": 5.1593,
      "mean_token_accuracy": 0.19939583390951157,
      "num_tokens": 91017169.0,
      "step": 43860
    },
    {
      "entropy": 5.491530561447144,
      "epoch": 3.979045718432511,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034960460035502117,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20650805681943893,
      "num_tokens": 91026684.0,
      "step": 43865
    },
    {
      "entropy": 5.452697229385376,
      "epoch": 3.9794992743105952,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003495740736147508,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2132559671998024,
      "num_tokens": 91037462.0,
      "step": 43870
    },
    {
      "entropy": 5.508552932739258,
      "epoch": 3.9799528301886795,
      "grad_norm": 0.875,
      "learning_rate": 0.00034954354533226953,
      "loss": 5.093,
      "mean_token_accuracy": 0.20319564640522003,
      "num_tokens": 91049418.0,
      "step": 43875
    },
    {
      "entropy": 5.4815136909484865,
      "epoch": 3.9804063860667633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003495130155082087,
      "loss": 4.9595,
      "mean_token_accuracy": 0.21141203194856645,
      "num_tokens": 91061013.0,
      "step": 43880
    },
    {
      "entropy": 5.460799264907837,
      "epoch": 3.9808599419448476,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003494824841431998,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21289751976728438,
      "num_tokens": 91070974.0,
      "step": 43885
    },
    {
      "entropy": 5.491414022445679,
      "epoch": 3.981313497822932,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003494519512378741,
      "loss": 5.0689,
      "mean_token_accuracy": 0.20940889418125153,
      "num_tokens": 91082681.0,
      "step": 43890
    },
    {
      "entropy": 5.495747423171997,
      "epoch": 3.9817670537010157,
      "grad_norm": 0.953125,
      "learning_rate": 0.00034942141679286307,
      "loss": 5.0127,
      "mean_token_accuracy": 0.2159772291779518,
      "num_tokens": 91093552.0,
      "step": 43895
    },
    {
      "entropy": 5.52347207069397,
      "epoch": 3.9822206095791,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003493908808087982,
      "loss": 5.068,
      "mean_token_accuracy": 0.20228153765201567,
      "num_tokens": 91105470.0,
      "step": 43900
    },
    {
      "entropy": 5.4810363292694095,
      "epoch": 3.9826741654571842,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000349360343286311,
      "loss": 5.0041,
      "mean_token_accuracy": 0.2134080559015274,
      "num_tokens": 91115822.0,
      "step": 43905
    },
    {
      "entropy": 5.450135374069214,
      "epoch": 3.9831277213352685,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034932980422603283,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21522330194711686,
      "num_tokens": 91125722.0,
      "step": 43910
    },
    {
      "entropy": 5.597023057937622,
      "epoch": 3.9835812772133528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003492992636285955,
      "loss": 5.1679,
      "mean_token_accuracy": 0.21152954846620559,
      "num_tokens": 91135783.0,
      "step": 43915
    },
    {
      "entropy": 5.527936029434204,
      "epoch": 3.984034833091437,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003492687214946304,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21648243367671965,
      "num_tokens": 91146151.0,
      "step": 43920
    },
    {
      "entropy": 5.487696552276612,
      "epoch": 3.9844883889695213,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00034923817782476927,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20494465827941893,
      "num_tokens": 91156711.0,
      "step": 43925
    },
    {
      "entropy": 5.476046991348267,
      "epoch": 3.984941944847605,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003492076326196436,
      "loss": 5.0541,
      "mean_token_accuracy": 0.20490773618221284,
      "num_tokens": 91166601.0,
      "step": 43930
    },
    {
      "entropy": 5.546845197677612,
      "epoch": 3.9853955007256894,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034917708587988523,
      "loss": 5.0773,
      "mean_token_accuracy": 0.2026481434702873,
      "num_tokens": 91176556.0,
      "step": 43935
    },
    {
      "entropy": 5.434735679626465,
      "epoch": 3.9858490566037736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034914653760612584,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21284545212984085,
      "num_tokens": 91186499.0,
      "step": 43940
    },
    {
      "entropy": 5.448588752746582,
      "epoch": 3.986302612481858,
      "grad_norm": 1.0,
      "learning_rate": 0.0003491159877989971,
      "loss": 4.9765,
      "mean_token_accuracy": 0.19908455908298492,
      "num_tokens": 91197088.0,
      "step": 43945
    },
    {
      "entropy": 5.49901180267334,
      "epoch": 3.9867561683599417,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000349085436459131,
      "loss": 5.0113,
      "mean_token_accuracy": 0.21499515026807786,
      "num_tokens": 91207188.0,
      "step": 43950
    },
    {
      "entropy": 5.476389741897583,
      "epoch": 3.987209724238026,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034905488358715907,
      "loss": 4.9926,
      "mean_token_accuracy": 0.2141385093331337,
      "num_tokens": 91217349.0,
      "step": 43955
    },
    {
      "entropy": 5.469604015350342,
      "epoch": 3.9876632801161103,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003490243291837133,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20822375416755676,
      "num_tokens": 91227775.0,
      "step": 43960
    },
    {
      "entropy": 5.478895092010498,
      "epoch": 3.9881168359941945,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034899377324942556,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22137465924024582,
      "num_tokens": 91238315.0,
      "step": 43965
    },
    {
      "entropy": 5.488349866867066,
      "epoch": 3.988570391872279,
      "grad_norm": 0.921875,
      "learning_rate": 0.00034896321578492767,
      "loss": 5.1271,
      "mean_token_accuracy": 0.20170524567365647,
      "num_tokens": 91249178.0,
      "step": 43970
    },
    {
      "entropy": 5.453508043289185,
      "epoch": 3.989023947750363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003489326567908516,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21149035692214965,
      "num_tokens": 91257911.0,
      "step": 43975
    },
    {
      "entropy": 5.501496505737305,
      "epoch": 3.989477503628447,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034890209626782943,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21348126381635665,
      "num_tokens": 91267491.0,
      "step": 43980
    },
    {
      "entropy": 5.447028350830078,
      "epoch": 3.989931059506531,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003488715342164931,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20977669358253478,
      "num_tokens": 91278292.0,
      "step": 43985
    },
    {
      "entropy": 5.530818271636963,
      "epoch": 3.9903846153846154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034884097063747453,
      "loss": 5.115,
      "mean_token_accuracy": 0.20079374313354492,
      "num_tokens": 91287871.0,
      "step": 43990
    },
    {
      "entropy": 5.4819803714752195,
      "epoch": 3.9908381712626997,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003488104055314059,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21859606355428696,
      "num_tokens": 91297201.0,
      "step": 43995
    },
    {
      "entropy": 5.3934898853302,
      "epoch": 3.9912917271407835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003487798388989193,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2205292046070099,
      "num_tokens": 91306449.0,
      "step": 44000
    },
    {
      "entropy": 5.389135074615479,
      "epoch": 3.9917452830188678,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034874927074064684,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21294040530920028,
      "num_tokens": 91317516.0,
      "step": 44005
    },
    {
      "entropy": 5.502745723724365,
      "epoch": 3.992198838896952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003487187010572206,
      "loss": 5.0601,
      "mean_token_accuracy": 0.20604586154222487,
      "num_tokens": 91326850.0,
      "step": 44010
    },
    {
      "entropy": 5.499790287017822,
      "epoch": 3.9926523947750363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003486881298492729,
      "loss": 5.0174,
      "mean_token_accuracy": 0.2067309021949768,
      "num_tokens": 91338068.0,
      "step": 44015
    },
    {
      "entropy": 5.544311666488648,
      "epoch": 3.9931059506531206,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003486575571174359,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2051831752061844,
      "num_tokens": 91349627.0,
      "step": 44020
    },
    {
      "entropy": 5.5029816150665285,
      "epoch": 3.993559506531205,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034862698286234185,
      "loss": 5.0788,
      "mean_token_accuracy": 0.20176617354154586,
      "num_tokens": 91361229.0,
      "step": 44025
    },
    {
      "entropy": 5.530025911331177,
      "epoch": 3.9940130624092887,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003485964070846229,
      "loss": 5.0944,
      "mean_token_accuracy": 0.20183354318141938,
      "num_tokens": 91372348.0,
      "step": 44030
    },
    {
      "entropy": 5.464128208160401,
      "epoch": 3.994466618287373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003485658297849117,
      "loss": 4.9572,
      "mean_token_accuracy": 0.2102286159992218,
      "num_tokens": 91383028.0,
      "step": 44035
    },
    {
      "entropy": 5.523563051223755,
      "epoch": 3.994920174165457,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034853525096384034,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20288071781396866,
      "num_tokens": 91393263.0,
      "step": 44040
    },
    {
      "entropy": 5.537549304962158,
      "epoch": 3.9953737300435415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003485046706220411,
      "loss": 5.0319,
      "mean_token_accuracy": 0.2135932445526123,
      "num_tokens": 91403881.0,
      "step": 44045
    },
    {
      "entropy": 5.4593220233917235,
      "epoch": 3.9958272859216253,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003484740887601467,
      "loss": 5.0086,
      "mean_token_accuracy": 0.211004139482975,
      "num_tokens": 91413723.0,
      "step": 44050
    },
    {
      "entropy": 5.489715099334717,
      "epoch": 3.9962808417997095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034844350537878937,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21007478684186937,
      "num_tokens": 91424168.0,
      "step": 44055
    },
    {
      "entropy": 5.577733421325684,
      "epoch": 3.996734397677794,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034841292047860164,
      "loss": 5.0859,
      "mean_token_accuracy": 0.2035103738307953,
      "num_tokens": 91433133.0,
      "step": 44060
    },
    {
      "entropy": 5.541143417358398,
      "epoch": 3.997187953555878,
      "grad_norm": 1.046875,
      "learning_rate": 0.000348382334060216,
      "loss": 5.1293,
      "mean_token_accuracy": 0.196157968044281,
      "num_tokens": 91444521.0,
      "step": 44065
    },
    {
      "entropy": 5.423428106307983,
      "epoch": 3.9976415094339623,
      "grad_norm": 1.09375,
      "learning_rate": 0.000348351746124265,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21396056562662125,
      "num_tokens": 91454870.0,
      "step": 44070
    },
    {
      "entropy": 5.471642732620239,
      "epoch": 3.9980950653120466,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003483211566713812,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21614213138818741,
      "num_tokens": 91465472.0,
      "step": 44075
    },
    {
      "entropy": 5.515135908126831,
      "epoch": 3.998548621190131,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003482905657021971,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20436827540397645,
      "num_tokens": 91475467.0,
      "step": 44080
    },
    {
      "entropy": 5.528747797012329,
      "epoch": 3.9990021770682147,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003482599732173456,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20965433716773987,
      "num_tokens": 91485615.0,
      "step": 44085
    },
    {
      "entropy": 5.463318490982056,
      "epoch": 3.999455732946299,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003482293792174591,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21416819840669632,
      "num_tokens": 91495790.0,
      "step": 44090
    },
    {
      "entropy": 5.51272177696228,
      "epoch": 3.9999092888243832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003481987837031703,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20405451208353043,
      "num_tokens": 91507164.0,
      "step": 44095
    },
    {
      "entropy": 5.467181253433227,
      "epoch": 4.000362844702467,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003481681866751121,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21930023282766342,
      "num_tokens": 91516732.0,
      "step": 44100
    },
    {
      "entropy": 5.386494779586792,
      "epoch": 4.000816400580551,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034813758813391716,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22184014767408372,
      "num_tokens": 91527073.0,
      "step": 44105
    },
    {
      "entropy": 5.409944677352906,
      "epoch": 4.001269956458636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003481069880802183,
      "loss": 4.934,
      "mean_token_accuracy": 0.21694376170635224,
      "num_tokens": 91538314.0,
      "step": 44110
    },
    {
      "entropy": 5.488721227645874,
      "epoch": 4.00172351233672,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034807638651464815,
      "loss": 4.97,
      "mean_token_accuracy": 0.21091042906045915,
      "num_tokens": 91549159.0,
      "step": 44115
    },
    {
      "entropy": 5.465715408325195,
      "epoch": 4.002177068214804,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003480457834378398,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20778342485427856,
      "num_tokens": 91560901.0,
      "step": 44120
    },
    {
      "entropy": 5.481712293624878,
      "epoch": 4.002630624092888,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034801517885042597,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21210503578186035,
      "num_tokens": 91570797.0,
      "step": 44125
    },
    {
      "entropy": 5.535243797302246,
      "epoch": 4.003084179970973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003479845727530397,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2043528139591217,
      "num_tokens": 91580971.0,
      "step": 44130
    },
    {
      "entropy": 5.424395942687989,
      "epoch": 4.003537735849057,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034795396514631386,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2160145476460457,
      "num_tokens": 91590834.0,
      "step": 44135
    },
    {
      "entropy": 5.453636217117309,
      "epoch": 4.003991291727141,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034792335603088135,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20862104296684264,
      "num_tokens": 91602099.0,
      "step": 44140
    },
    {
      "entropy": 5.453024530410767,
      "epoch": 4.004444847605225,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034789274540737535,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2224446564912796,
      "num_tokens": 91612269.0,
      "step": 44145
    },
    {
      "entropy": 5.479214668273926,
      "epoch": 4.004898403483309,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034786213327642874,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21338489949703215,
      "num_tokens": 91623644.0,
      "step": 44150
    },
    {
      "entropy": 5.43748140335083,
      "epoch": 4.005351959361393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003478315196386747,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2178834080696106,
      "num_tokens": 91633563.0,
      "step": 44155
    },
    {
      "entropy": 5.458177661895752,
      "epoch": 4.005805515239477,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003478009044947463,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21607141941785812,
      "num_tokens": 91644648.0,
      "step": 44160
    },
    {
      "entropy": 5.428993129730225,
      "epoch": 4.006259071117562,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003477702878452766,
      "loss": 4.841,
      "mean_token_accuracy": 0.23899659365415574,
      "num_tokens": 91655366.0,
      "step": 44165
    },
    {
      "entropy": 5.402198791503906,
      "epoch": 4.006712626995646,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003477396696908988,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21801353693008424,
      "num_tokens": 91665474.0,
      "step": 44170
    },
    {
      "entropy": 5.44046983718872,
      "epoch": 4.00716618287373,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034770905003224606,
      "loss": 4.926,
      "mean_token_accuracy": 0.21318466663360597,
      "num_tokens": 91675291.0,
      "step": 44175
    },
    {
      "entropy": 5.365434885025024,
      "epoch": 4.007619738751814,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003476784288699517,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22023553401231766,
      "num_tokens": 91686472.0,
      "step": 44180
    },
    {
      "entropy": 5.464283275604248,
      "epoch": 4.008073294629899,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034764780620464886,
      "loss": 4.9796,
      "mean_token_accuracy": 0.21344097405672074,
      "num_tokens": 91696347.0,
      "step": 44185
    },
    {
      "entropy": 5.460143184661865,
      "epoch": 4.008526850507983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003476171820369709,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20822201818227767,
      "num_tokens": 91706504.0,
      "step": 44190
    },
    {
      "entropy": 5.422375345230103,
      "epoch": 4.008980406386066,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034758655636755103,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22339455038309097,
      "num_tokens": 91716559.0,
      "step": 44195
    },
    {
      "entropy": 5.478626823425293,
      "epoch": 4.009433962264151,
      "grad_norm": 1.125,
      "learning_rate": 0.00034755592919702275,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20935356318950654,
      "num_tokens": 91726312.0,
      "step": 44200
    },
    {
      "entropy": 5.43790864944458,
      "epoch": 4.009887518142235,
      "grad_norm": 0.953125,
      "learning_rate": 0.00034752530052601944,
      "loss": 4.8791,
      "mean_token_accuracy": 0.21419404298067093,
      "num_tokens": 91736882.0,
      "step": 44205
    },
    {
      "entropy": 5.466473722457886,
      "epoch": 4.010341074020319,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034749467035517437,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21067962646484376,
      "num_tokens": 91746905.0,
      "step": 44210
    },
    {
      "entropy": 5.443706750869751,
      "epoch": 4.010794629898403,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034746403868512094,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20603534132242202,
      "num_tokens": 91757609.0,
      "step": 44215
    },
    {
      "entropy": 5.466469764709473,
      "epoch": 4.011248185776488,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034743340551649285,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22114092856645584,
      "num_tokens": 91768017.0,
      "step": 44220
    },
    {
      "entropy": 5.416880893707275,
      "epoch": 4.011701741654572,
      "grad_norm": 1.0,
      "learning_rate": 0.00034740277084992346,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22493513822555541,
      "num_tokens": 91777820.0,
      "step": 44225
    },
    {
      "entropy": 5.382226181030274,
      "epoch": 4.012155297532656,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034737213468604626,
      "loss": 4.9453,
      "mean_token_accuracy": 0.2118930160999298,
      "num_tokens": 91789011.0,
      "step": 44230
    },
    {
      "entropy": 5.497543144226074,
      "epoch": 4.0126088534107405,
      "grad_norm": 1.125,
      "learning_rate": 0.0003473414970254949,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2117552012205124,
      "num_tokens": 91799165.0,
      "step": 44235
    },
    {
      "entropy": 5.459438800811768,
      "epoch": 4.013062409288825,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034731085786890306,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21844685822725296,
      "num_tokens": 91809602.0,
      "step": 44240
    },
    {
      "entropy": 5.481701946258545,
      "epoch": 4.013515965166909,
      "grad_norm": 1.25,
      "learning_rate": 0.00034728021721690407,
      "loss": 5.0045,
      "mean_token_accuracy": 0.21175531148910523,
      "num_tokens": 91820092.0,
      "step": 44245
    },
    {
      "entropy": 5.449554586410523,
      "epoch": 4.013969521044992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003472495750701318,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21952653527259827,
      "num_tokens": 91829730.0,
      "step": 44250
    },
    {
      "entropy": 5.500666379928589,
      "epoch": 4.014423076923077,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034721893142922,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2072078064084053,
      "num_tokens": 91840697.0,
      "step": 44255
    },
    {
      "entropy": 5.460888910293579,
      "epoch": 4.014876632801161,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003471882862948023,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2074793115258217,
      "num_tokens": 91852145.0,
      "step": 44260
    },
    {
      "entropy": 5.471689653396607,
      "epoch": 4.015330188679245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034715763966751233,
      "loss": 4.9128,
      "mean_token_accuracy": 0.20639771670103074,
      "num_tokens": 91861940.0,
      "step": 44265
    },
    {
      "entropy": 5.4997295379638675,
      "epoch": 4.0157837445573294,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034712699154798403,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21546147167682647,
      "num_tokens": 91872179.0,
      "step": 44270
    },
    {
      "entropy": 5.47741060256958,
      "epoch": 4.016237300435414,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034709634193685126,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21825528889894485,
      "num_tokens": 91882051.0,
      "step": 44275
    },
    {
      "entropy": 5.410479211807251,
      "epoch": 4.016690856313498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003470656908347476,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20844220221042634,
      "num_tokens": 91892672.0,
      "step": 44280
    },
    {
      "entropy": 5.396726608276367,
      "epoch": 4.017144412191582,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003470350382423072,
      "loss": 4.8578,
      "mean_token_accuracy": 0.22088557481765747,
      "num_tokens": 91903418.0,
      "step": 44285
    },
    {
      "entropy": 5.433375263214112,
      "epoch": 4.0175979680696665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003470043841601639,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21678256690502168,
      "num_tokens": 91913076.0,
      "step": 44290
    },
    {
      "entropy": 5.437510299682617,
      "epoch": 4.018051523947751,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034697372858895147,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22067774832248688,
      "num_tokens": 91923464.0,
      "step": 44295
    },
    {
      "entropy": 5.515762424468994,
      "epoch": 4.018505079825834,
      "grad_norm": 1.09375,
      "learning_rate": 0.000346943071529304,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20839064121246337,
      "num_tokens": 91934399.0,
      "step": 44300
    },
    {
      "entropy": 5.533348131179809,
      "epoch": 4.018958635703918,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034691241298185553,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2150438129901886,
      "num_tokens": 91945330.0,
      "step": 44305
    },
    {
      "entropy": 5.462036418914795,
      "epoch": 4.019412191582003,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034688175294724004,
      "loss": 4.9806,
      "mean_token_accuracy": 0.2094249501824379,
      "num_tokens": 91956379.0,
      "step": 44310
    },
    {
      "entropy": 5.313155794143677,
      "epoch": 4.019865747460087,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034685109142609154,
      "loss": 4.8103,
      "mean_token_accuracy": 0.2227373778820038,
      "num_tokens": 91965969.0,
      "step": 44315
    },
    {
      "entropy": 5.513428258895874,
      "epoch": 4.020319303338171,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003468204284190442,
      "loss": 5.0747,
      "mean_token_accuracy": 0.19997566044330597,
      "num_tokens": 91977005.0,
      "step": 44320
    },
    {
      "entropy": 5.548929500579834,
      "epoch": 4.0207728592162555,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034678976392673206,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21650888919830322,
      "num_tokens": 91988735.0,
      "step": 44325
    },
    {
      "entropy": 5.50764045715332,
      "epoch": 4.02122641509434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034675909794978926,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21275002062320708,
      "num_tokens": 91998743.0,
      "step": 44330
    },
    {
      "entropy": 5.449676942825318,
      "epoch": 4.021679970972424,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034672843048885017,
      "loss": 4.98,
      "mean_token_accuracy": 0.2138497933745384,
      "num_tokens": 92009465.0,
      "step": 44335
    },
    {
      "entropy": 5.353665447235107,
      "epoch": 4.022133526850508,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034669776154454875,
      "loss": 4.9068,
      "mean_token_accuracy": 0.22203533053398133,
      "num_tokens": 92020818.0,
      "step": 44340
    },
    {
      "entropy": 5.504662036895752,
      "epoch": 4.0225870827285926,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034666709111751945,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20908742547035217,
      "num_tokens": 92031528.0,
      "step": 44345
    },
    {
      "entropy": 5.587652540206909,
      "epoch": 4.023040638606676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003466364192083963,
      "loss": 5.019,
      "mean_token_accuracy": 0.21078341603279113,
      "num_tokens": 92041856.0,
      "step": 44350
    },
    {
      "entropy": 5.498630428314209,
      "epoch": 4.02349419448476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034660574581781386,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21968589574098588,
      "num_tokens": 92052117.0,
      "step": 44355
    },
    {
      "entropy": 5.462101840972901,
      "epoch": 4.0239477503628445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003465750709464063,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2094442993402481,
      "num_tokens": 92062405.0,
      "step": 44360
    },
    {
      "entropy": 5.468239641189575,
      "epoch": 4.024401306240929,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000346544394594808,
      "loss": 5.0144,
      "mean_token_accuracy": 0.21452679485082626,
      "num_tokens": 92073269.0,
      "step": 44365
    },
    {
      "entropy": 5.454206466674805,
      "epoch": 4.024854862119013,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034651371676365343,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2236509993672371,
      "num_tokens": 92083645.0,
      "step": 44370
    },
    {
      "entropy": 5.568469476699829,
      "epoch": 4.025308417997097,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034648303745357705,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21458503603935242,
      "num_tokens": 92093696.0,
      "step": 44375
    },
    {
      "entropy": 5.4479306697845455,
      "epoch": 4.0257619738751815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003464523566652131,
      "loss": 4.9782,
      "mean_token_accuracy": 0.205788654088974,
      "num_tokens": 92103217.0,
      "step": 44380
    },
    {
      "entropy": 5.378258180618286,
      "epoch": 4.026215529753266,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034642167439919634,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21486997306346894,
      "num_tokens": 92112934.0,
      "step": 44385
    },
    {
      "entropy": 5.4430845260620115,
      "epoch": 4.02666908563135,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003463909906561611,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21876087188720703,
      "num_tokens": 92123066.0,
      "step": 44390
    },
    {
      "entropy": 5.467815732955932,
      "epoch": 4.027122641509434,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003463603054367419,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21956997364759445,
      "num_tokens": 92133911.0,
      "step": 44395
    },
    {
      "entropy": 5.355214214324951,
      "epoch": 4.027576197387518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003463296187415735,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2180871531367302,
      "num_tokens": 92143941.0,
      "step": 44400
    },
    {
      "entropy": 5.416090583801269,
      "epoch": 4.028029753265602,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003462989305712904,
      "loss": 4.8992,
      "mean_token_accuracy": 0.21751432865858078,
      "num_tokens": 92153934.0,
      "step": 44405
    },
    {
      "entropy": 5.457577180862427,
      "epoch": 4.028483309143686,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003462682409265272,
      "loss": 4.88,
      "mean_token_accuracy": 0.2206941142678261,
      "num_tokens": 92163642.0,
      "step": 44410
    },
    {
      "entropy": 5.440938329696655,
      "epoch": 4.0289368650217705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003462375498079186,
      "loss": 4.9592,
      "mean_token_accuracy": 0.2129038989543915,
      "num_tokens": 92174034.0,
      "step": 44415
    },
    {
      "entropy": 5.491748952865601,
      "epoch": 4.029390420899855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034620685721609937,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2153406947851181,
      "num_tokens": 92183714.0,
      "step": 44420
    },
    {
      "entropy": 5.459010314941406,
      "epoch": 4.029843976777939,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00034617616315170413,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20657615959644318,
      "num_tokens": 92195291.0,
      "step": 44425
    },
    {
      "entropy": 5.4797553539276125,
      "epoch": 4.030297532656023,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003461454676153678,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21346121579408645,
      "num_tokens": 92206159.0,
      "step": 44430
    },
    {
      "entropy": 5.499883985519409,
      "epoch": 4.030751088534108,
      "grad_norm": 1.09375,
      "learning_rate": 0.000346114770607725,
      "loss": 5.0117,
      "mean_token_accuracy": 0.21007079482078553,
      "num_tokens": 92216928.0,
      "step": 44435
    },
    {
      "entropy": 5.3959300994873045,
      "epoch": 4.031204644412192,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003460840721294106,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22052426040172576,
      "num_tokens": 92227029.0,
      "step": 44440
    },
    {
      "entropy": 5.448106575012207,
      "epoch": 4.031658200290276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034605337218105955,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21610728949308394,
      "num_tokens": 92237310.0,
      "step": 44445
    },
    {
      "entropy": 5.425592613220215,
      "epoch": 4.03211175616836,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003460226707633066,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22079939246177674,
      "num_tokens": 92247841.0,
      "step": 44450
    },
    {
      "entropy": 5.443043518066406,
      "epoch": 4.032565312046444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034599196787678667,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21030958145856857,
      "num_tokens": 92258513.0,
      "step": 44455
    },
    {
      "entropy": 5.46356987953186,
      "epoch": 4.033018867924528,
      "grad_norm": 1.125,
      "learning_rate": 0.0003459612635221349,
      "loss": 4.9636,
      "mean_token_accuracy": 0.2120138570666313,
      "num_tokens": 92268659.0,
      "step": 44460
    },
    {
      "entropy": 5.4592485427856445,
      "epoch": 4.033472423802612,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034593055769998595,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21153529435396196,
      "num_tokens": 92279810.0,
      "step": 44465
    },
    {
      "entropy": 5.481690168380737,
      "epoch": 4.0339259796806965,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003458998504109751,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2175609812140465,
      "num_tokens": 92289625.0,
      "step": 44470
    },
    {
      "entropy": 5.402948474884033,
      "epoch": 4.034379535558781,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003458691416557372,
      "loss": 4.9089,
      "mean_token_accuracy": 0.2115829646587372,
      "num_tokens": 92300669.0,
      "step": 44475
    },
    {
      "entropy": 5.430788326263428,
      "epoch": 4.034833091436865,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034583843143490744,
      "loss": 4.9769,
      "mean_token_accuracy": 0.21860619485378266,
      "num_tokens": 92312160.0,
      "step": 44480
    },
    {
      "entropy": 5.535433149337768,
      "epoch": 4.035286647314949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003458077197491208,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2037322461605072,
      "num_tokens": 92322677.0,
      "step": 44485
    },
    {
      "entropy": 5.4649515628814695,
      "epoch": 4.035740203193034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003457770065990126,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21939283311367036,
      "num_tokens": 92333148.0,
      "step": 44490
    },
    {
      "entropy": 5.479809284210205,
      "epoch": 4.036193759071118,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003457462919852178,
      "loss": 4.998,
      "mean_token_accuracy": 0.21101005226373673,
      "num_tokens": 92344189.0,
      "step": 44495
    },
    {
      "entropy": 5.5196675777435305,
      "epoch": 4.036647314949202,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003457155759083717,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21517525017261505,
      "num_tokens": 92356136.0,
      "step": 44500
    },
    {
      "entropy": 5.4199591159820555,
      "epoch": 4.0371008708272855,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003456848583691094,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22038530111312865,
      "num_tokens": 92366393.0,
      "step": 44505
    },
    {
      "entropy": 5.427608680725098,
      "epoch": 4.03755442670537,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003456541393680662,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21917283535003662,
      "num_tokens": 92376819.0,
      "step": 44510
    },
    {
      "entropy": 5.423156785964966,
      "epoch": 4.038007982583454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034562341890587747,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21160225570201874,
      "num_tokens": 92386704.0,
      "step": 44515
    },
    {
      "entropy": 5.466674852371216,
      "epoch": 4.038461538461538,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003455926969831784,
      "loss": 5.0023,
      "mean_token_accuracy": 0.21045105159282684,
      "num_tokens": 92397412.0,
      "step": 44520
    },
    {
      "entropy": 5.578925228118896,
      "epoch": 4.038915094339623,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034556197360060436,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2075452610850334,
      "num_tokens": 92408641.0,
      "step": 44525
    },
    {
      "entropy": 5.45826096534729,
      "epoch": 4.039368650217707,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003455312487587908,
      "loss": 4.9346,
      "mean_token_accuracy": 0.22057772129774095,
      "num_tokens": 92419341.0,
      "step": 44530
    },
    {
      "entropy": 5.444120073318482,
      "epoch": 4.039822206095791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003455005224583729,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21466944217681885,
      "num_tokens": 92429297.0,
      "step": 44535
    },
    {
      "entropy": 5.505938291549683,
      "epoch": 4.040275761973875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034546979469998637,
      "loss": 5.0025,
      "mean_token_accuracy": 0.21340746581554412,
      "num_tokens": 92439851.0,
      "step": 44540
    },
    {
      "entropy": 5.3855311393737795,
      "epoch": 4.04072931785196,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034543906548426646,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21722103506326676,
      "num_tokens": 92451096.0,
      "step": 44545
    },
    {
      "entropy": 5.48360686302185,
      "epoch": 4.041182873730044,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034540833481184867,
      "loss": 4.9152,
      "mean_token_accuracy": 0.218068665266037,
      "num_tokens": 92460516.0,
      "step": 44550
    },
    {
      "entropy": 5.566098928451538,
      "epoch": 4.041636429608127,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034537760268336866,
      "loss": 4.999,
      "mean_token_accuracy": 0.20959622263908387,
      "num_tokens": 92471429.0,
      "step": 44555
    },
    {
      "entropy": 5.482111978530884,
      "epoch": 4.042089985486212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003453468690994619,
      "loss": 4.9456,
      "mean_token_accuracy": 0.21687395870685577,
      "num_tokens": 92482047.0,
      "step": 44560
    },
    {
      "entropy": 5.393990755081177,
      "epoch": 4.042543541364296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034531613406076387,
      "loss": 4.8741,
      "mean_token_accuracy": 0.21877649128437043,
      "num_tokens": 92492206.0,
      "step": 44565
    },
    {
      "entropy": 5.372890758514404,
      "epoch": 4.04299709724238,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003452853975679103,
      "loss": 4.856,
      "mean_token_accuracy": 0.21520789116621017,
      "num_tokens": 92501979.0,
      "step": 44570
    },
    {
      "entropy": 5.570201396942139,
      "epoch": 4.043450653120464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034525465962153683,
      "loss": 5.0266,
      "mean_token_accuracy": 0.21146403402090072,
      "num_tokens": 92512388.0,
      "step": 44575
    },
    {
      "entropy": 5.471607875823975,
      "epoch": 4.043904208998549,
      "grad_norm": 1.125,
      "learning_rate": 0.00034522392022227915,
      "loss": 5.0014,
      "mean_token_accuracy": 0.21046460866928102,
      "num_tokens": 92522805.0,
      "step": 44580
    },
    {
      "entropy": 5.44204421043396,
      "epoch": 4.044357764876633,
      "grad_norm": 1.0,
      "learning_rate": 0.00034519317937077283,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2072525814175606,
      "num_tokens": 92532957.0,
      "step": 44585
    },
    {
      "entropy": 5.494346046447754,
      "epoch": 4.044811320754717,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003451624370676537,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20763078778982164,
      "num_tokens": 92543451.0,
      "step": 44590
    },
    {
      "entropy": 5.547148036956787,
      "epoch": 4.045264876632801,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034513169331355754,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2134159117937088,
      "num_tokens": 92553856.0,
      "step": 44595
    },
    {
      "entropy": 5.351771545410156,
      "epoch": 4.045718432510886,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034510094810911995,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22069500833749772,
      "num_tokens": 92563816.0,
      "step": 44600
    },
    {
      "entropy": 5.420434236526489,
      "epoch": 4.04617198838897,
      "grad_norm": 1.0,
      "learning_rate": 0.00034507020145497705,
      "loss": 4.9573,
      "mean_token_accuracy": 0.21748400777578353,
      "num_tokens": 92574062.0,
      "step": 44605
    },
    {
      "entropy": 5.424419975280761,
      "epoch": 4.046625544267053,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034503945335176444,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22410139292478562,
      "num_tokens": 92583935.0,
      "step": 44610
    },
    {
      "entropy": 5.509253263473511,
      "epoch": 4.047079100145138,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034500870380011807,
      "loss": 4.961,
      "mean_token_accuracy": 0.21653029918670655,
      "num_tokens": 92594037.0,
      "step": 44615
    },
    {
      "entropy": 5.4361536502838135,
      "epoch": 4.047532656023222,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034497795280067386,
      "loss": 4.902,
      "mean_token_accuracy": 0.21153976917266845,
      "num_tokens": 92603336.0,
      "step": 44620
    },
    {
      "entropy": 5.363025808334351,
      "epoch": 4.047986211901306,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034494720035406777,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22250016480684282,
      "num_tokens": 92614210.0,
      "step": 44625
    },
    {
      "entropy": 5.468615961074829,
      "epoch": 4.04843976777939,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003449164464609358,
      "loss": 4.9556,
      "mean_token_accuracy": 0.214266075193882,
      "num_tokens": 92624036.0,
      "step": 44630
    },
    {
      "entropy": 5.492751407623291,
      "epoch": 4.048893323657475,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003448856911219139,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20701131373643875,
      "num_tokens": 92634588.0,
      "step": 44635
    },
    {
      "entropy": 5.428782606124878,
      "epoch": 4.049346879535559,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034485493433763804,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2166955813765526,
      "num_tokens": 92645027.0,
      "step": 44640
    },
    {
      "entropy": 5.538189792633057,
      "epoch": 4.049800435413643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034482417610874437,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21235018223524094,
      "num_tokens": 92656669.0,
      "step": 44645
    },
    {
      "entropy": 5.469654321670532,
      "epoch": 4.0502539912917275,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003447934164358689,
      "loss": 5.0377,
      "mean_token_accuracy": 0.21035591065883635,
      "num_tokens": 92666558.0,
      "step": 44650
    },
    {
      "entropy": 5.491444206237793,
      "epoch": 4.050707547169812,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034476265531964786,
      "loss": 4.9856,
      "mean_token_accuracy": 0.2158972755074501,
      "num_tokens": 92677384.0,
      "step": 44655
    },
    {
      "entropy": 5.411927556991577,
      "epoch": 4.051161103047895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003447318927607173,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2192441090941429,
      "num_tokens": 92687619.0,
      "step": 44660
    },
    {
      "entropy": 5.375108861923218,
      "epoch": 4.051614658925979,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003447011287597134,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22508492916822434,
      "num_tokens": 92697784.0,
      "step": 44665
    },
    {
      "entropy": 5.462611389160156,
      "epoch": 4.052068214804064,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003446703633172724,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2062777265906334,
      "num_tokens": 92709255.0,
      "step": 44670
    },
    {
      "entropy": 5.431706094741822,
      "epoch": 4.052521770682148,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003446395964340305,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22004402726888656,
      "num_tokens": 92720714.0,
      "step": 44675
    },
    {
      "entropy": 5.453010368347168,
      "epoch": 4.052975326560232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034460882811062417,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21622052639722825,
      "num_tokens": 92730783.0,
      "step": 44680
    },
    {
      "entropy": 5.487659454345703,
      "epoch": 4.0534288824383164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034457805834768937,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20703331232070923,
      "num_tokens": 92741943.0,
      "step": 44685
    },
    {
      "entropy": 5.427710676193238,
      "epoch": 4.053882438316401,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003445472871458627,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21635450869798661,
      "num_tokens": 92753510.0,
      "step": 44690
    },
    {
      "entropy": 5.442355728149414,
      "epoch": 4.054335994194485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003445165145057803,
      "loss": 4.9023,
      "mean_token_accuracy": 0.21834689527750015,
      "num_tokens": 92763655.0,
      "step": 44695
    },
    {
      "entropy": 5.5390218734741214,
      "epoch": 4.054789550072569,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034448574042807866,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21262889951467515,
      "num_tokens": 92773732.0,
      "step": 44700
    },
    {
      "entropy": 5.482279634475708,
      "epoch": 4.0552431059506535,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003444549649133943,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21210775524377823,
      "num_tokens": 92784822.0,
      "step": 44705
    },
    {
      "entropy": 5.496907520294189,
      "epoch": 4.055696661828737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003444241879623636,
      "loss": 5.0488,
      "mean_token_accuracy": 0.19923135936260222,
      "num_tokens": 92795270.0,
      "step": 44710
    },
    {
      "entropy": 5.473844575881958,
      "epoch": 4.056150217706821,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003443934095756229,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21318014115095138,
      "num_tokens": 92806578.0,
      "step": 44715
    },
    {
      "entropy": 5.483394479751587,
      "epoch": 4.056603773584905,
      "grad_norm": 1.0,
      "learning_rate": 0.0003443626297538088,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21364026218652726,
      "num_tokens": 92815932.0,
      "step": 44720
    },
    {
      "entropy": 5.388333082199097,
      "epoch": 4.05705732946299,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003443318484975579,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21958848237991332,
      "num_tokens": 92825536.0,
      "step": 44725
    },
    {
      "entropy": 5.446198987960815,
      "epoch": 4.057510885341074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003443010658075066,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21504178643226624,
      "num_tokens": 92836451.0,
      "step": 44730
    },
    {
      "entropy": 5.495935440063477,
      "epoch": 4.057964441219158,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003442702816842917,
      "loss": 5.045,
      "mean_token_accuracy": 0.20900655239820481,
      "num_tokens": 92846923.0,
      "step": 44735
    },
    {
      "entropy": 5.458924198150635,
      "epoch": 4.0584179970972425,
      "grad_norm": 0.953125,
      "learning_rate": 0.00034423949612854974,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21932675838470458,
      "num_tokens": 92857549.0,
      "step": 44740
    },
    {
      "entropy": 5.495053005218506,
      "epoch": 4.058871552975327,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003442087091409173,
      "loss": 5.0067,
      "mean_token_accuracy": 0.2087995409965515,
      "num_tokens": 92867059.0,
      "step": 44745
    },
    {
      "entropy": 5.453888845443726,
      "epoch": 4.059325108853411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003441779207220311,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20439936965703964,
      "num_tokens": 92878138.0,
      "step": 44750
    },
    {
      "entropy": 5.475457000732422,
      "epoch": 4.059778664731495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003441471308725279,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21236603260040282,
      "num_tokens": 92888931.0,
      "step": 44755
    },
    {
      "entropy": 5.504412317276001,
      "epoch": 4.060232220609579,
      "grad_norm": 1.0,
      "learning_rate": 0.00034411633959304444,
      "loss": 5.0182,
      "mean_token_accuracy": 0.2059352368116379,
      "num_tokens": 92899718.0,
      "step": 44760
    },
    {
      "entropy": 5.49577226638794,
      "epoch": 4.060685776487663,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00034408554688421743,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2203524276614189,
      "num_tokens": 92910381.0,
      "step": 44765
    },
    {
      "entropy": 5.448528146743774,
      "epoch": 4.061139332365747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003440547527466837,
      "loss": 4.9148,
      "mean_token_accuracy": 0.2180999383330345,
      "num_tokens": 92919905.0,
      "step": 44770
    },
    {
      "entropy": 5.434178733825684,
      "epoch": 4.0615928882438315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034402395718108016,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21780400723218918,
      "num_tokens": 92930210.0,
      "step": 44775
    },
    {
      "entropy": 5.433261108398438,
      "epoch": 4.062046444121916,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034399316018804356,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21584614515304565,
      "num_tokens": 92940832.0,
      "step": 44780
    },
    {
      "entropy": 5.532689428329467,
      "epoch": 4.0625,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003439623617682107,
      "loss": 5.0285,
      "mean_token_accuracy": 0.20841735750436782,
      "num_tokens": 92951349.0,
      "step": 44785
    },
    {
      "entropy": 5.469904947280884,
      "epoch": 4.062953555878084,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003439315619222188,
      "loss": 4.9475,
      "mean_token_accuracy": 0.21073243767023087,
      "num_tokens": 92961642.0,
      "step": 44790
    },
    {
      "entropy": 5.466363000869751,
      "epoch": 4.0634071117561685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003439007606507046,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2190684735774994,
      "num_tokens": 92971061.0,
      "step": 44795
    },
    {
      "entropy": 5.507037591934204,
      "epoch": 4.063860667634253,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003438699579543051,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20386780202388763,
      "num_tokens": 92981732.0,
      "step": 44800
    },
    {
      "entropy": 5.496320629119873,
      "epoch": 4.064314223512337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034383915383365726,
      "loss": 5.044,
      "mean_token_accuracy": 0.2093049868941307,
      "num_tokens": 92991570.0,
      "step": 44805
    },
    {
      "entropy": 5.506445455551147,
      "epoch": 4.064767779390421,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003438083482893982,
      "loss": 4.9664,
      "mean_token_accuracy": 0.21644256860017777,
      "num_tokens": 93001623.0,
      "step": 44810
    },
    {
      "entropy": 5.50038743019104,
      "epoch": 4.065221335268505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034377754132216496,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21102894693613053,
      "num_tokens": 93012167.0,
      "step": 44815
    },
    {
      "entropy": 5.413044691085815,
      "epoch": 4.065674891146589,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003437467329325947,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22783767133951188,
      "num_tokens": 93022090.0,
      "step": 44820
    },
    {
      "entropy": 5.421819067001342,
      "epoch": 4.066128447024673,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003437159231213244,
      "loss": 4.827,
      "mean_token_accuracy": 0.22642624527215957,
      "num_tokens": 93030917.0,
      "step": 44825
    },
    {
      "entropy": 5.4456291675567625,
      "epoch": 4.0665820029027575,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003436851118889914,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21325236558914185,
      "num_tokens": 93040113.0,
      "step": 44830
    },
    {
      "entropy": 5.494983386993408,
      "epoch": 4.067035558780842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003436542992362326,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21804309338331224,
      "num_tokens": 93050053.0,
      "step": 44835
    },
    {
      "entropy": 5.558561754226685,
      "epoch": 4.067489114658926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003436234851636856,
      "loss": 5.031,
      "mean_token_accuracy": 0.20578498095273973,
      "num_tokens": 93060581.0,
      "step": 44840
    },
    {
      "entropy": 5.512717914581299,
      "epoch": 4.06794267053701,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034359266967198734,
      "loss": 5.0125,
      "mean_token_accuracy": 0.21495127975940703,
      "num_tokens": 93070158.0,
      "step": 44845
    },
    {
      "entropy": 5.3504773616790775,
      "epoch": 4.068396226415095,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003435618527617752,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21954845488071442,
      "num_tokens": 93080804.0,
      "step": 44850
    },
    {
      "entropy": 5.451083755493164,
      "epoch": 4.068849782293179,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003435310344336865,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21423922777175902,
      "num_tokens": 93091346.0,
      "step": 44855
    },
    {
      "entropy": 5.421881008148193,
      "epoch": 4.069303338171263,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003435002146883585,
      "loss": 4.9009,
      "mean_token_accuracy": 0.219980388879776,
      "num_tokens": 93100931.0,
      "step": 44860
    },
    {
      "entropy": 5.44166579246521,
      "epoch": 4.0697568940493465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003434693935264286,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21984275430440903,
      "num_tokens": 93110181.0,
      "step": 44865
    },
    {
      "entropy": 5.469074535369873,
      "epoch": 4.070210449927431,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003434385709485343,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21563574969768523,
      "num_tokens": 93120140.0,
      "step": 44870
    },
    {
      "entropy": 5.514381742477417,
      "epoch": 4.070664005805515,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003434077469553128,
      "loss": 4.9779,
      "mean_token_accuracy": 0.20729571878910064,
      "num_tokens": 93130853.0,
      "step": 44875
    },
    {
      "entropy": 5.478063917160034,
      "epoch": 4.071117561683599,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003433769215474017,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2189699739217758,
      "num_tokens": 93141430.0,
      "step": 44880
    },
    {
      "entropy": 5.499154853820801,
      "epoch": 4.0715711175616836,
      "grad_norm": 1.0,
      "learning_rate": 0.00034334609472543845,
      "loss": 5.063,
      "mean_token_accuracy": 0.20631178915500642,
      "num_tokens": 93152270.0,
      "step": 44885
    },
    {
      "entropy": 5.5180481433868405,
      "epoch": 4.072024673439768,
      "grad_norm": 1.125,
      "learning_rate": 0.0003433152664900605,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20903230607509612,
      "num_tokens": 93162850.0,
      "step": 44890
    },
    {
      "entropy": 5.488230752944946,
      "epoch": 4.072478229317852,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003432844368419055,
      "loss": 4.8325,
      "mean_token_accuracy": 0.22301695197820665,
      "num_tokens": 93172265.0,
      "step": 44895
    },
    {
      "entropy": 5.406287336349488,
      "epoch": 4.072931785195936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003432536057816108,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22243899554014207,
      "num_tokens": 93182756.0,
      "step": 44900
    },
    {
      "entropy": 5.4166967391967775,
      "epoch": 4.073385341074021,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034322277330981426,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21064148992300033,
      "num_tokens": 93192710.0,
      "step": 44905
    },
    {
      "entropy": 5.419774913787842,
      "epoch": 4.073838896952105,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034319193942715343,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21902046501636505,
      "num_tokens": 93202531.0,
      "step": 44910
    },
    {
      "entropy": 5.528845977783203,
      "epoch": 4.074292452830188,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034316110413426586,
      "loss": 5.0098,
      "mean_token_accuracy": 0.21224775314331054,
      "num_tokens": 93212208.0,
      "step": 44915
    },
    {
      "entropy": 5.56564359664917,
      "epoch": 4.0747460087082725,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003431302674317892,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20529954135417938,
      "num_tokens": 93222932.0,
      "step": 44920
    },
    {
      "entropy": 5.4363807201385494,
      "epoch": 4.075199564586357,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034309942932036136,
      "loss": 4.8231,
      "mean_token_accuracy": 0.2266455501317978,
      "num_tokens": 93232049.0,
      "step": 44925
    },
    {
      "entropy": 5.484938097000122,
      "epoch": 4.075653120464441,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034306858980061994,
      "loss": 5.0166,
      "mean_token_accuracy": 0.21265263706445695,
      "num_tokens": 93244230.0,
      "step": 44930
    },
    {
      "entropy": 5.442882680892945,
      "epoch": 4.076106676342525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034303774887320274,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21323266327381135,
      "num_tokens": 93254159.0,
      "step": 44935
    },
    {
      "entropy": 5.432450437545777,
      "epoch": 4.07656023222061,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034300690653874753,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22550283074378968,
      "num_tokens": 93263742.0,
      "step": 44940
    },
    {
      "entropy": 5.458405637741089,
      "epoch": 4.077013788098694,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00034297606279789215,
      "loss": 5.0473,
      "mean_token_accuracy": 0.2057817578315735,
      "num_tokens": 93274446.0,
      "step": 44945
    },
    {
      "entropy": 5.457405138015747,
      "epoch": 4.077467343976778,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003429452176512745,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20756917893886567,
      "num_tokens": 93284102.0,
      "step": 44950
    },
    {
      "entropy": 5.4734913349151615,
      "epoch": 4.077920899854862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034291437109953244,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22306899577379227,
      "num_tokens": 93295096.0,
      "step": 44955
    },
    {
      "entropy": 5.461998319625854,
      "epoch": 4.078374455732947,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003428835231433039,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21165355741977693,
      "num_tokens": 93305377.0,
      "step": 44960
    },
    {
      "entropy": 5.588665962219238,
      "epoch": 4.07882801161103,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034285267378322674,
      "loss": 5.0462,
      "mean_token_accuracy": 0.21237242668867112,
      "num_tokens": 93315859.0,
      "step": 44965
    },
    {
      "entropy": 5.480920743942261,
      "epoch": 4.079281567489114,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00034282182301993903,
      "loss": 4.92,
      "mean_token_accuracy": 0.2158252015709877,
      "num_tokens": 93326519.0,
      "step": 44970
    },
    {
      "entropy": 5.464319515228271,
      "epoch": 4.079735123367199,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003427909708540787,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21300482004880905,
      "num_tokens": 93337055.0,
      "step": 44975
    },
    {
      "entropy": 5.420218420028687,
      "epoch": 4.080188679245283,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034276011728628393,
      "loss": 5.0098,
      "mean_token_accuracy": 0.21040676534175873,
      "num_tokens": 93346565.0,
      "step": 44980
    },
    {
      "entropy": 5.461569118499756,
      "epoch": 4.080642235123367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034272926231719255,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21464691460132598,
      "num_tokens": 93356366.0,
      "step": 44985
    },
    {
      "entropy": 5.53213210105896,
      "epoch": 4.081095791001451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003426984059474428,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21038717776536942,
      "num_tokens": 93367317.0,
      "step": 44990
    },
    {
      "entropy": 5.4823661804199215,
      "epoch": 4.081549346879536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003426675481776728,
      "loss": 4.9818,
      "mean_token_accuracy": 0.2103107303380966,
      "num_tokens": 93379601.0,
      "step": 44995
    },
    {
      "entropy": 5.436238145828247,
      "epoch": 4.08200290275762,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034263668900852065,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2238496348261833,
      "num_tokens": 93389740.0,
      "step": 45000
    },
    {
      "epoch": 4.08200290275762,
      "eval_entropy": 5.249903142506181,
      "eval_loss": 5.075412750244141,
      "eval_mean_token_accuracy": 0.21600360598465707,
      "eval_num_tokens": 93389740.0,
      "eval_runtime": 26.6712,
      "eval_samples_per_second": 1325.773,
      "eval_steps_per_second": 165.722,
      "step": 45000
    },
    {
      "entropy": 5.567705678939819,
      "epoch": 4.082456458635704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003426058284406244,
      "loss": 5.0775,
      "mean_token_accuracy": 0.20762218981981279,
      "num_tokens": 93400232.0,
      "step": 45005
    },
    {
      "entropy": 5.487520551681518,
      "epoch": 4.082910014513788,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003425749664746226,
      "loss": 5.0049,
      "mean_token_accuracy": 0.21525383442640306,
      "num_tokens": 93410662.0,
      "step": 45010
    },
    {
      "entropy": 5.4131368637084964,
      "epoch": 4.083363570391873,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003425441031111532,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22524733394384383,
      "num_tokens": 93420328.0,
      "step": 45015
    },
    {
      "entropy": 5.461620664596557,
      "epoch": 4.083817126269956,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034251323835085453,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20430761575698853,
      "num_tokens": 93430878.0,
      "step": 45020
    },
    {
      "entropy": 5.591705179214477,
      "epoch": 4.08427068214804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003424823721943649,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20949413180351256,
      "num_tokens": 93441784.0,
      "step": 45025
    },
    {
      "entropy": 5.5015185356140135,
      "epoch": 4.084724238026125,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003424515046423226,
      "loss": 4.9894,
      "mean_token_accuracy": 0.21614145189523698,
      "num_tokens": 93453058.0,
      "step": 45030
    },
    {
      "entropy": 5.48646354675293,
      "epoch": 4.085177793904209,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003424206356953659,
      "loss": 4.9974,
      "mean_token_accuracy": 0.21253278255462646,
      "num_tokens": 93463089.0,
      "step": 45035
    },
    {
      "entropy": 5.452635002136231,
      "epoch": 4.085631349782293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034238976535413333,
      "loss": 4.9813,
      "mean_token_accuracy": 0.22038712054491044,
      "num_tokens": 93473315.0,
      "step": 45040
    },
    {
      "entropy": 5.5821065425872805,
      "epoch": 4.086084905660377,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003423588936192633,
      "loss": 5.0595,
      "mean_token_accuracy": 0.19818170070648194,
      "num_tokens": 93485113.0,
      "step": 45045
    },
    {
      "entropy": 5.426789712905884,
      "epoch": 4.086538461538462,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003423280204913941,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21635770052671432,
      "num_tokens": 93494269.0,
      "step": 45050
    },
    {
      "entropy": 5.348138952255249,
      "epoch": 4.086992017416546,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003422971459711644,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22356011718511581,
      "num_tokens": 93505052.0,
      "step": 45055
    },
    {
      "entropy": 5.439647197723389,
      "epoch": 4.08744557329463,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034226627005921244,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21822881400585176,
      "num_tokens": 93514986.0,
      "step": 45060
    },
    {
      "entropy": 5.478863620758057,
      "epoch": 4.0878991291727145,
      "grad_norm": 1.109375,
      "learning_rate": 0.000342235392756177,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2121971607208252,
      "num_tokens": 93525286.0,
      "step": 45065
    },
    {
      "entropy": 5.499686908721924,
      "epoch": 4.088352685050798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034220451406269636,
      "loss": 4.979,
      "mean_token_accuracy": 0.2137551039457321,
      "num_tokens": 93536034.0,
      "step": 45070
    },
    {
      "entropy": 5.402642059326172,
      "epoch": 4.088806240928882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034217363397940937,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2190207615494728,
      "num_tokens": 93546283.0,
      "step": 45075
    },
    {
      "entropy": 5.427524852752685,
      "epoch": 4.089259796806966,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034214275250695446,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21974662244319915,
      "num_tokens": 93555811.0,
      "step": 45080
    },
    {
      "entropy": 5.545194625854492,
      "epoch": 4.089713352685051,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003421118696459703,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20676959156990052,
      "num_tokens": 93567154.0,
      "step": 45085
    },
    {
      "entropy": 5.454268550872802,
      "epoch": 4.090166908563135,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034208098539709566,
      "loss": 4.9171,
      "mean_token_accuracy": 0.22617148905992507,
      "num_tokens": 93577654.0,
      "step": 45090
    },
    {
      "entropy": 5.4084227085113525,
      "epoch": 4.090620464441219,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034205009976096906,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21748674660921097,
      "num_tokens": 93588712.0,
      "step": 45095
    },
    {
      "entropy": 5.44348554611206,
      "epoch": 4.0910740203193035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003420192127382293,
      "loss": 5.0017,
      "mean_token_accuracy": 0.20789168328046798,
      "num_tokens": 93599186.0,
      "step": 45100
    },
    {
      "entropy": 5.4312177181243895,
      "epoch": 4.091527576197388,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034198832432951517,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22718443721532822,
      "num_tokens": 93609439.0,
      "step": 45105
    },
    {
      "entropy": 5.4322387218475345,
      "epoch": 4.091981132075472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003419574345354654,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21074725091457366,
      "num_tokens": 93619903.0,
      "step": 45110
    },
    {
      "entropy": 5.465011358261108,
      "epoch": 4.092434687953556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034192654335671893,
      "loss": 4.8729,
      "mean_token_accuracy": 0.2222719222307205,
      "num_tokens": 93629895.0,
      "step": 45115
    },
    {
      "entropy": 5.447215890884399,
      "epoch": 4.09288824383164,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034189565079391436,
      "loss": 5.0182,
      "mean_token_accuracy": 0.21173955351114274,
      "num_tokens": 93640677.0,
      "step": 45120
    },
    {
      "entropy": 5.492579603195191,
      "epoch": 4.093341799709724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003418647568476907,
      "loss": 5.0513,
      "mean_token_accuracy": 0.21378433853387832,
      "num_tokens": 93651541.0,
      "step": 45125
    },
    {
      "entropy": 5.551086950302124,
      "epoch": 4.093795355587808,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003418338615186869,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20827688425779342,
      "num_tokens": 93661901.0,
      "step": 45130
    },
    {
      "entropy": 5.478535842895508,
      "epoch": 4.094248911465892,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034180296480754174,
      "loss": 4.9537,
      "mean_token_accuracy": 0.21460937112569808,
      "num_tokens": 93672307.0,
      "step": 45135
    },
    {
      "entropy": 5.4511946678161625,
      "epoch": 4.094702467343977,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003417720667148943,
      "loss": 4.9087,
      "mean_token_accuracy": 0.206627781689167,
      "num_tokens": 93683802.0,
      "step": 45140
    },
    {
      "entropy": 5.523177242279052,
      "epoch": 4.095156023222061,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003417411672413835,
      "loss": 5.0346,
      "mean_token_accuracy": 0.2064689353108406,
      "num_tokens": 93693467.0,
      "step": 45145
    },
    {
      "entropy": 5.44119701385498,
      "epoch": 4.095609579100145,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034171026638764833,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2170169711112976,
      "num_tokens": 93703858.0,
      "step": 45150
    },
    {
      "entropy": 5.462725830078125,
      "epoch": 4.0960631349782295,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034167936415432785,
      "loss": 4.9626,
      "mean_token_accuracy": 0.2244664177298546,
      "num_tokens": 93714457.0,
      "step": 45155
    },
    {
      "entropy": 5.440717458724976,
      "epoch": 4.096516690856314,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003416484605420611,
      "loss": 4.966,
      "mean_token_accuracy": 0.21848649382591248,
      "num_tokens": 93725009.0,
      "step": 45160
    },
    {
      "entropy": 5.542855072021484,
      "epoch": 4.096970246734398,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00034161755555148725,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20454669892787933,
      "num_tokens": 93736357.0,
      "step": 45165
    },
    {
      "entropy": 5.466707468032837,
      "epoch": 4.097423802612482,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00034158664918324536,
      "loss": 4.919,
      "mean_token_accuracy": 0.21113361120224,
      "num_tokens": 93746711.0,
      "step": 45170
    },
    {
      "entropy": 5.495560503005981,
      "epoch": 4.097877358490566,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00034155574143797455,
      "loss": 5.032,
      "mean_token_accuracy": 0.20764389932155608,
      "num_tokens": 93757732.0,
      "step": 45175
    },
    {
      "entropy": 5.453432893753051,
      "epoch": 4.09833091436865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034152483231631413,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21448816508054733,
      "num_tokens": 93767048.0,
      "step": 45180
    },
    {
      "entropy": 5.606799554824829,
      "epoch": 4.098784470246734,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003414939218189032,
      "loss": 5.153,
      "mean_token_accuracy": 0.20314402133226395,
      "num_tokens": 93778283.0,
      "step": 45185
    },
    {
      "entropy": 5.514069032669068,
      "epoch": 4.0992380261248185,
      "grad_norm": 1.0625,
      "learning_rate": 0.000341463009946381,
      "loss": 5.06,
      "mean_token_accuracy": 0.21016367375850678,
      "num_tokens": 93788776.0,
      "step": 45190
    },
    {
      "entropy": 5.481219005584717,
      "epoch": 4.099691582002903,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034143209669938685,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20992327779531478,
      "num_tokens": 93798695.0,
      "step": 45195
    },
    {
      "entropy": 5.550480365753174,
      "epoch": 4.100145137880987,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034140118207856,
      "loss": 5.0611,
      "mean_token_accuracy": 0.21531792283058165,
      "num_tokens": 93808993.0,
      "step": 45200
    },
    {
      "entropy": 5.557712602615356,
      "epoch": 4.100598693759071,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034137026608453973,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20950680822134018,
      "num_tokens": 93818361.0,
      "step": 45205
    },
    {
      "entropy": 5.540270376205444,
      "epoch": 4.1010522496371555,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003413393487179655,
      "loss": 4.9534,
      "mean_token_accuracy": 0.22354677319526672,
      "num_tokens": 93828988.0,
      "step": 45210
    },
    {
      "entropy": 5.5694067001342775,
      "epoch": 4.10150580551524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003413084299794767,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20967113524675368,
      "num_tokens": 93839350.0,
      "step": 45215
    },
    {
      "entropy": 5.490941429138184,
      "epoch": 4.101959361393324,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034127750986971255,
      "loss": 4.9439,
      "mean_token_accuracy": 0.218784661591053,
      "num_tokens": 93849939.0,
      "step": 45220
    },
    {
      "entropy": 5.48854660987854,
      "epoch": 4.1024129172714074,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003412465883893126,
      "loss": 4.9928,
      "mean_token_accuracy": 0.21124818176031113,
      "num_tokens": 93861278.0,
      "step": 45225
    },
    {
      "entropy": 5.510129356384278,
      "epoch": 4.102866473149492,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003412156655389164,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21041972041130066,
      "num_tokens": 93871693.0,
      "step": 45230
    },
    {
      "entropy": 5.484918546676636,
      "epoch": 4.103320029027576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003411847413191633,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21418211311101915,
      "num_tokens": 93883098.0,
      "step": 45235
    },
    {
      "entropy": 5.42204704284668,
      "epoch": 4.10377358490566,
      "grad_norm": 1.0,
      "learning_rate": 0.00034115381573069297,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21803656965494156,
      "num_tokens": 93893347.0,
      "step": 45240
    },
    {
      "entropy": 5.514180660247803,
      "epoch": 4.1042271407837445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034112288877414485,
      "loss": 5.0032,
      "mean_token_accuracy": 0.20449741780757905,
      "num_tokens": 93903964.0,
      "step": 45245
    },
    {
      "entropy": 5.507415437698365,
      "epoch": 4.104680696661829,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003410919604501585,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21657335758209229,
      "num_tokens": 93914526.0,
      "step": 45250
    },
    {
      "entropy": 5.539132308959961,
      "epoch": 4.105134252539913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003410610307593735,
      "loss": 5.0788,
      "mean_token_accuracy": 0.2097782552242279,
      "num_tokens": 93924430.0,
      "step": 45255
    },
    {
      "entropy": 5.492255544662475,
      "epoch": 4.105587808417997,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003410300997024297,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21776138693094255,
      "num_tokens": 93935150.0,
      "step": 45260
    },
    {
      "entropy": 5.422422981262207,
      "epoch": 4.106041364296082,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003409991672799665,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2202281430363655,
      "num_tokens": 93944496.0,
      "step": 45265
    },
    {
      "entropy": 5.481423616409302,
      "epoch": 4.106494920174166,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034096823349262376,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2066260814666748,
      "num_tokens": 93955745.0,
      "step": 45270
    },
    {
      "entropy": 5.3955545902252195,
      "epoch": 4.106948476052249,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003409372983410411,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21195199638605117,
      "num_tokens": 93965930.0,
      "step": 45275
    },
    {
      "entropy": 5.438071393966675,
      "epoch": 4.1074020319303335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003409063618258583,
      "loss": 4.8897,
      "mean_token_accuracy": 0.21776217073202134,
      "num_tokens": 93976435.0,
      "step": 45280
    },
    {
      "entropy": 5.380533456802368,
      "epoch": 4.107855587808418,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034087542394771524,
      "loss": 4.8154,
      "mean_token_accuracy": 0.22384563237428665,
      "num_tokens": 93987294.0,
      "step": 45285
    },
    {
      "entropy": 5.51188440322876,
      "epoch": 4.108309143686502,
      "grad_norm": 1.125,
      "learning_rate": 0.00034084448470725147,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2100375309586525,
      "num_tokens": 93997177.0,
      "step": 45290
    },
    {
      "entropy": 5.5450758934021,
      "epoch": 4.108762699564586,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003408135441051071,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20635457783937455,
      "num_tokens": 94007663.0,
      "step": 45295
    },
    {
      "entropy": 5.444271183013916,
      "epoch": 4.1092162554426706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003407826021419218,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2168340042233467,
      "num_tokens": 94016899.0,
      "step": 45300
    },
    {
      "entropy": 5.469272613525391,
      "epoch": 4.109669811320755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003407516588183356,
      "loss": 5.0382,
      "mean_token_accuracy": 0.2079993173480034,
      "num_tokens": 94027696.0,
      "step": 45305
    },
    {
      "entropy": 5.416297721862793,
      "epoch": 4.110123367198839,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003407207141349882,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21254483759403228,
      "num_tokens": 94038075.0,
      "step": 45310
    },
    {
      "entropy": 5.501027679443359,
      "epoch": 4.110576923076923,
      "grad_norm": 1.0,
      "learning_rate": 0.0003406897680925199,
      "loss": 4.9912,
      "mean_token_accuracy": 0.21047068983316422,
      "num_tokens": 94049082.0,
      "step": 45315
    },
    {
      "entropy": 5.501435470581055,
      "epoch": 4.111030478955008,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034065882069157023,
      "loss": 4.988,
      "mean_token_accuracy": 0.2107749506831169,
      "num_tokens": 94060367.0,
      "step": 45320
    },
    {
      "entropy": 5.468975973129273,
      "epoch": 4.111484034833092,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034062787193277954,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2174751028418541,
      "num_tokens": 94069415.0,
      "step": 45325
    },
    {
      "entropy": 5.475853538513183,
      "epoch": 4.111937590711175,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00034059692181678775,
      "loss": 4.977,
      "mean_token_accuracy": 0.20851955860853194,
      "num_tokens": 94079047.0,
      "step": 45330
    },
    {
      "entropy": 5.443176317214966,
      "epoch": 4.1123911465892595,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034056597034423484,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21611464619636536,
      "num_tokens": 94089631.0,
      "step": 45335
    },
    {
      "entropy": 5.466036128997803,
      "epoch": 4.112844702467344,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034053501751576097,
      "loss": 4.918,
      "mean_token_accuracy": 0.2157600238919258,
      "num_tokens": 94099265.0,
      "step": 45340
    },
    {
      "entropy": 5.433970594406128,
      "epoch": 4.113298258345428,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003405040633320063,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21131886839866637,
      "num_tokens": 94109673.0,
      "step": 45345
    },
    {
      "entropy": 5.4140465259552,
      "epoch": 4.113751814223512,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003404731077936108,
      "loss": 4.854,
      "mean_token_accuracy": 0.22638339996337892,
      "num_tokens": 94120004.0,
      "step": 45350
    },
    {
      "entropy": 5.378581714630127,
      "epoch": 4.114205370101597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003404421509012148,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2142721563577652,
      "num_tokens": 94129935.0,
      "step": 45355
    },
    {
      "entropy": 5.47610993385315,
      "epoch": 4.114658925979681,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034041119265545847,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21146164536476136,
      "num_tokens": 94139035.0,
      "step": 45360
    },
    {
      "entropy": 5.516679048538208,
      "epoch": 4.115112481857765,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034038023305698196,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21083794683218002,
      "num_tokens": 94149281.0,
      "step": 45365
    },
    {
      "entropy": 5.4582761287689205,
      "epoch": 4.115566037735849,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003403492721064256,
      "loss": 4.945,
      "mean_token_accuracy": 0.21416784077882767,
      "num_tokens": 94160455.0,
      "step": 45370
    },
    {
      "entropy": 5.506740283966065,
      "epoch": 4.116019593613934,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034031830980442955,
      "loss": 4.9934,
      "mean_token_accuracy": 0.21064430475234985,
      "num_tokens": 94170334.0,
      "step": 45375
    },
    {
      "entropy": 5.448571014404297,
      "epoch": 4.116473149492017,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034028734615163436,
      "loss": 4.8933,
      "mean_token_accuracy": 0.2165552020072937,
      "num_tokens": 94181055.0,
      "step": 45380
    },
    {
      "entropy": 5.45173978805542,
      "epoch": 4.116926705370101,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003402563811486802,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2192676618695259,
      "num_tokens": 94191182.0,
      "step": 45385
    },
    {
      "entropy": 5.463207530975342,
      "epoch": 4.117380261248186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003402254147962073,
      "loss": 5.0041,
      "mean_token_accuracy": 0.21080698519945146,
      "num_tokens": 94201229.0,
      "step": 45390
    },
    {
      "entropy": 5.414521598815918,
      "epoch": 4.11783381712627,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003401944470948563,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21571021378040314,
      "num_tokens": 94210956.0,
      "step": 45395
    },
    {
      "entropy": 5.474539566040039,
      "epoch": 4.118287373004354,
      "grad_norm": 1.0,
      "learning_rate": 0.00034016347804526753,
      "loss": 4.9642,
      "mean_token_accuracy": 0.21288579404354097,
      "num_tokens": 94220740.0,
      "step": 45400
    },
    {
      "entropy": 5.414825344085694,
      "epoch": 4.118740928882438,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00034013250764808133,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2311682790517807,
      "num_tokens": 94231497.0,
      "step": 45405
    },
    {
      "entropy": 5.462934160232544,
      "epoch": 4.119194484760523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003401015359039384,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21756968945264815,
      "num_tokens": 94240492.0,
      "step": 45410
    },
    {
      "entropy": 5.414997005462647,
      "epoch": 4.119648040638607,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003400705628134791,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2154642254114151,
      "num_tokens": 94251586.0,
      "step": 45415
    },
    {
      "entropy": 5.437297058105469,
      "epoch": 4.120101596516691,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000340039588377344,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2199668139219284,
      "num_tokens": 94261635.0,
      "step": 45420
    },
    {
      "entropy": 5.446309328079224,
      "epoch": 4.120555152394775,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003400086125961735,
      "loss": 4.987,
      "mean_token_accuracy": 0.2099283292889595,
      "num_tokens": 94269881.0,
      "step": 45425
    },
    {
      "entropy": 5.428301095962524,
      "epoch": 4.121008708272859,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033997763547060847,
      "loss": 4.92,
      "mean_token_accuracy": 0.21534616202116014,
      "num_tokens": 94279412.0,
      "step": 45430
    },
    {
      "entropy": 5.480091714859009,
      "epoch": 4.121462264150943,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003399466570012893,
      "loss": 4.993,
      "mean_token_accuracy": 0.21457578539848327,
      "num_tokens": 94290157.0,
      "step": 45435
    },
    {
      "entropy": 5.520541191101074,
      "epoch": 4.121915820029027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033991567718885674,
      "loss": 4.988,
      "mean_token_accuracy": 0.21303920000791549,
      "num_tokens": 94299693.0,
      "step": 45440
    },
    {
      "entropy": 5.5823382377624515,
      "epoch": 4.122369375907112,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003398846960339515,
      "loss": 5.0686,
      "mean_token_accuracy": 0.2090616136789322,
      "num_tokens": 94310245.0,
      "step": 45445
    },
    {
      "entropy": 5.452808570861817,
      "epoch": 4.122822931785196,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003398537135372141,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2109529420733452,
      "num_tokens": 94321190.0,
      "step": 45450
    },
    {
      "entropy": 5.498926496505737,
      "epoch": 4.12327648766328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003398227296992854,
      "loss": 5.0075,
      "mean_token_accuracy": 0.22176835238933562,
      "num_tokens": 94330273.0,
      "step": 45455
    },
    {
      "entropy": 5.537972259521484,
      "epoch": 4.123730043541364,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033979174452080616,
      "loss": 5.0569,
      "mean_token_accuracy": 0.20639471113681793,
      "num_tokens": 94341097.0,
      "step": 45460
    },
    {
      "entropy": 5.487576198577881,
      "epoch": 4.124183599419449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003397607580024171,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21193207204341888,
      "num_tokens": 94352329.0,
      "step": 45465
    },
    {
      "entropy": 5.4658928394317625,
      "epoch": 4.124637155297533,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003397297701447591,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2135487198829651,
      "num_tokens": 94362551.0,
      "step": 45470
    },
    {
      "entropy": 5.454349327087402,
      "epoch": 4.125090711175617,
      "grad_norm": 1.046875,
      "learning_rate": 0.000339698780948473,
      "loss": 5.0099,
      "mean_token_accuracy": 0.2108696922659874,
      "num_tokens": 94372405.0,
      "step": 45475
    },
    {
      "entropy": 5.538887596130371,
      "epoch": 4.125544267053701,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003396677904141995,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21613918691873552,
      "num_tokens": 94382347.0,
      "step": 45480
    },
    {
      "entropy": 5.417104768753052,
      "epoch": 4.125997822931785,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00033963679854257965,
      "loss": 4.9793,
      "mean_token_accuracy": 0.2117331564426422,
      "num_tokens": 94393428.0,
      "step": 45485
    },
    {
      "entropy": 5.4802836894989015,
      "epoch": 4.126451378809869,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003396058053342544,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2108530357480049,
      "num_tokens": 94403941.0,
      "step": 45490
    },
    {
      "entropy": 5.494494819641114,
      "epoch": 4.126904934687953,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033957481078986465,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21126089245080948,
      "num_tokens": 94414206.0,
      "step": 45495
    },
    {
      "entropy": 5.4778697967529295,
      "epoch": 4.127358490566038,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033954381491005126,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2167613163590431,
      "num_tokens": 94424481.0,
      "step": 45500
    },
    {
      "entropy": 5.435950088500976,
      "epoch": 4.127812046444122,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003395128176954554,
      "loss": 4.9126,
      "mean_token_accuracy": 0.22186049371957778,
      "num_tokens": 94434714.0,
      "step": 45505
    },
    {
      "entropy": 5.509081125259399,
      "epoch": 4.128265602322206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000339481819146718,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20977237820625305,
      "num_tokens": 94445267.0,
      "step": 45510
    },
    {
      "entropy": 5.501987552642822,
      "epoch": 4.1287191582002905,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033945081926448013,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2103172242641449,
      "num_tokens": 94455724.0,
      "step": 45515
    },
    {
      "entropy": 5.536451148986816,
      "epoch": 4.129172714078375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033941981804938294,
      "loss": 5.0269,
      "mean_token_accuracy": 0.21077206283807753,
      "num_tokens": 94466655.0,
      "step": 45520
    },
    {
      "entropy": 5.445908403396606,
      "epoch": 4.129626269956459,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033938881550206744,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22070303559303284,
      "num_tokens": 94477699.0,
      "step": 45525
    },
    {
      "entropy": 5.483239698410034,
      "epoch": 4.130079825834542,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033935781162317485,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21712489873170854,
      "num_tokens": 94487639.0,
      "step": 45530
    },
    {
      "entropy": 5.56452431678772,
      "epoch": 4.130533381712627,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033932680641334635,
      "loss": 5.0869,
      "mean_token_accuracy": 0.20544599294662474,
      "num_tokens": 94498687.0,
      "step": 45535
    },
    {
      "entropy": 5.461418390274048,
      "epoch": 4.130986937590711,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003392957998732231,
      "loss": 4.9198,
      "mean_token_accuracy": 0.22160128504037857,
      "num_tokens": 94508706.0,
      "step": 45540
    },
    {
      "entropy": 5.3952723979949955,
      "epoch": 4.131440493468795,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033926479200344626,
      "loss": 4.8552,
      "mean_token_accuracy": 0.227299328148365,
      "num_tokens": 94517246.0,
      "step": 45545
    },
    {
      "entropy": 5.420139408111572,
      "epoch": 4.131894049346879,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003392337828046572,
      "loss": 4.87,
      "mean_token_accuracy": 0.21837302148342133,
      "num_tokens": 94527956.0,
      "step": 45550
    },
    {
      "entropy": 5.535203218460083,
      "epoch": 4.132347605224964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003392027722774971,
      "loss": 5.0428,
      "mean_token_accuracy": 0.20747627317905426,
      "num_tokens": 94539101.0,
      "step": 45555
    },
    {
      "entropy": 5.4593291759490965,
      "epoch": 4.132801161103048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033917176042260735,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21504084020853043,
      "num_tokens": 94549597.0,
      "step": 45560
    },
    {
      "entropy": 5.484008836746216,
      "epoch": 4.133254716981132,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003391407472406292,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20949673056602477,
      "num_tokens": 94559911.0,
      "step": 45565
    },
    {
      "entropy": 5.458158254623413,
      "epoch": 4.1337082728592165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033910973273220416,
      "loss": 4.8897,
      "mean_token_accuracy": 0.21729089617729186,
      "num_tokens": 94569396.0,
      "step": 45570
    },
    {
      "entropy": 5.4441815376281735,
      "epoch": 4.134161828737301,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033907871689797337,
      "loss": 5.0002,
      "mean_token_accuracy": 0.219846473634243,
      "num_tokens": 94579817.0,
      "step": 45575
    },
    {
      "entropy": 5.374127769470215,
      "epoch": 4.134615384615385,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003390476997385785,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2156280368566513,
      "num_tokens": 94590398.0,
      "step": 45580
    },
    {
      "entropy": 5.451160335540772,
      "epoch": 4.135068940493468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003390166812546608,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21450913846492767,
      "num_tokens": 94600528.0,
      "step": 45585
    },
    {
      "entropy": 5.484153938293457,
      "epoch": 4.135522496371553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033898566144686184,
      "loss": 4.9625,
      "mean_token_accuracy": 0.21296287924051285,
      "num_tokens": 94611107.0,
      "step": 45590
    },
    {
      "entropy": 5.5599524974823,
      "epoch": 4.135976052249637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003389546403158231,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20533659309148788,
      "num_tokens": 94620932.0,
      "step": 45595
    },
    {
      "entropy": 5.517901182174683,
      "epoch": 4.136429608127721,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033892361786218604,
      "loss": 5.0039,
      "mean_token_accuracy": 0.2077041521668434,
      "num_tokens": 94630476.0,
      "step": 45600
    },
    {
      "entropy": 5.439888668060303,
      "epoch": 4.1368831640058055,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003388925940865924,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21134912073612214,
      "num_tokens": 94641487.0,
      "step": 45605
    },
    {
      "entropy": 5.472664737701416,
      "epoch": 4.13733671988389,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00033886156898968353,
      "loss": 4.863,
      "mean_token_accuracy": 0.22417924404144288,
      "num_tokens": 94651668.0,
      "step": 45610
    },
    {
      "entropy": 5.382763433456421,
      "epoch": 4.137790275761974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033883054257210116,
      "loss": 4.9676,
      "mean_token_accuracy": 0.20880414843559264,
      "num_tokens": 94661218.0,
      "step": 45615
    },
    {
      "entropy": 5.4058661460876465,
      "epoch": 4.138243831640058,
      "grad_norm": 1.09375,
      "learning_rate": 0.000338799514834487,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21054814755916595,
      "num_tokens": 94672412.0,
      "step": 45620
    },
    {
      "entropy": 5.510491943359375,
      "epoch": 4.1386973875181425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003387684857774824,
      "loss": 4.9933,
      "mean_token_accuracy": 0.2084800586104393,
      "num_tokens": 94683102.0,
      "step": 45625
    },
    {
      "entropy": 5.479387283325195,
      "epoch": 4.139150943396227,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003387374554017294,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21887557655572892,
      "num_tokens": 94692869.0,
      "step": 45630
    },
    {
      "entropy": 5.465399169921875,
      "epoch": 4.13960449927431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033870642370786966,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2133125126361847,
      "num_tokens": 94702437.0,
      "step": 45635
    },
    {
      "entropy": 5.473424339294434,
      "epoch": 4.1400580551523944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033867539069654476,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2111661359667778,
      "num_tokens": 94713130.0,
      "step": 45640
    },
    {
      "entropy": 5.517959928512573,
      "epoch": 4.140511611030479,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033864435636839647,
      "loss": 5.043,
      "mean_token_accuracy": 0.2087673619389534,
      "num_tokens": 94723251.0,
      "step": 45645
    },
    {
      "entropy": 5.4571140766143795,
      "epoch": 4.140965166908563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003386133207240668,
      "loss": 4.9814,
      "mean_token_accuracy": 0.21973376274108886,
      "num_tokens": 94733805.0,
      "step": 45650
    },
    {
      "entropy": 5.451752185821533,
      "epoch": 4.141418722786647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003385822837641974,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21320863217115402,
      "num_tokens": 94743654.0,
      "step": 45655
    },
    {
      "entropy": 5.449963998794556,
      "epoch": 4.1418722786647315,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033855124548943025,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21268097460269927,
      "num_tokens": 94754635.0,
      "step": 45660
    },
    {
      "entropy": 5.513573408126831,
      "epoch": 4.142325834542816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033852020590040705,
      "loss": 5.0117,
      "mean_token_accuracy": 0.2134435772895813,
      "num_tokens": 94765371.0,
      "step": 45665
    },
    {
      "entropy": 5.433696603775024,
      "epoch": 4.1427793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003384891649977698,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22666950523853302,
      "num_tokens": 94775780.0,
      "step": 45670
    },
    {
      "entropy": 5.482482528686523,
      "epoch": 4.143232946298984,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033845812278216045,
      "loss": 4.9803,
      "mean_token_accuracy": 0.20764864087104798,
      "num_tokens": 94785345.0,
      "step": 45675
    },
    {
      "entropy": 5.415020942687988,
      "epoch": 4.143686502177069,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033842707925422095,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21953038424253463,
      "num_tokens": 94796059.0,
      "step": 45680
    },
    {
      "entropy": 5.449004983901977,
      "epoch": 4.144140058055152,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033839603441459333,
      "loss": 4.928,
      "mean_token_accuracy": 0.21911748498678207,
      "num_tokens": 94805867.0,
      "step": 45685
    },
    {
      "entropy": 5.417425966262817,
      "epoch": 4.144593613933236,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003383649882639195,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21400771290063858,
      "num_tokens": 94817172.0,
      "step": 45690
    },
    {
      "entropy": 5.454342269897461,
      "epoch": 4.1450471698113205,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003383339408028416,
      "loss": 4.952,
      "mean_token_accuracy": 0.2124848932027817,
      "num_tokens": 94828748.0,
      "step": 45695
    },
    {
      "entropy": 5.4591383934021,
      "epoch": 4.145500725689405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003383028920320017,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21983875930309296,
      "num_tokens": 94838704.0,
      "step": 45700
    },
    {
      "entropy": 5.531021022796631,
      "epoch": 4.145954281567489,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00033827184195204174,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20396330207586288,
      "num_tokens": 94849555.0,
      "step": 45705
    },
    {
      "entropy": 5.567774534225464,
      "epoch": 4.146407837445573,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003382407905636041,
      "loss": 5.0683,
      "mean_token_accuracy": 0.2044032856822014,
      "num_tokens": 94862238.0,
      "step": 45710
    },
    {
      "entropy": 5.36355767250061,
      "epoch": 4.1468613933236576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003382097378673308,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22472131699323655,
      "num_tokens": 94872960.0,
      "step": 45715
    },
    {
      "entropy": 5.3956122398376465,
      "epoch": 4.147314949201742,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000338178683863864,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21958278864622116,
      "num_tokens": 94883612.0,
      "step": 45720
    },
    {
      "entropy": 5.406961154937744,
      "epoch": 4.147768505079826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003381476285538458,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22561696469783782,
      "num_tokens": 94893176.0,
      "step": 45725
    },
    {
      "entropy": 5.421048831939697,
      "epoch": 4.14822206095791,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003381165719379187,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2124757409095764,
      "num_tokens": 94903860.0,
      "step": 45730
    },
    {
      "entropy": 5.413828992843628,
      "epoch": 4.148675616835995,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003380855140167247,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22115300595760345,
      "num_tokens": 94915681.0,
      "step": 45735
    },
    {
      "entropy": 5.471819067001343,
      "epoch": 4.149129172714078,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003380544547909061,
      "loss": 4.9317,
      "mean_token_accuracy": 0.214432255923748,
      "num_tokens": 94927062.0,
      "step": 45740
    },
    {
      "entropy": 5.45979266166687,
      "epoch": 4.149582728592162,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003380233942611055,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21821820586919785,
      "num_tokens": 94938838.0,
      "step": 45745
    },
    {
      "entropy": 5.410526275634766,
      "epoch": 4.1500362844702465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000337992332427965,
      "loss": 4.8938,
      "mean_token_accuracy": 0.21689582914113997,
      "num_tokens": 94949678.0,
      "step": 45750
    },
    {
      "entropy": 5.475602817535401,
      "epoch": 4.150489840348331,
      "grad_norm": 1.015625,
      "learning_rate": 0.000337961269292127,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21134773045778274,
      "num_tokens": 94960314.0,
      "step": 45755
    },
    {
      "entropy": 5.453370904922485,
      "epoch": 4.150943396226415,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033793020485423385,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21334167569875717,
      "num_tokens": 94969831.0,
      "step": 45760
    },
    {
      "entropy": 5.4971604347229,
      "epoch": 4.151396952104499,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033789913911492814,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21137058287858962,
      "num_tokens": 94980046.0,
      "step": 45765
    },
    {
      "entropy": 5.4227275371551515,
      "epoch": 4.151850507982584,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003378680720748522,
      "loss": 4.9596,
      "mean_token_accuracy": 0.20946562439203262,
      "num_tokens": 94990850.0,
      "step": 45770
    },
    {
      "entropy": 5.5378459930419925,
      "epoch": 4.152304063860668,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003378370037346484,
      "loss": 5.0077,
      "mean_token_accuracy": 0.21005648076534272,
      "num_tokens": 95001658.0,
      "step": 45775
    },
    {
      "entropy": 5.48968358039856,
      "epoch": 4.152757619738752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033780593409495936,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21797353625297547,
      "num_tokens": 95011912.0,
      "step": 45780
    },
    {
      "entropy": 5.423772382736206,
      "epoch": 4.153211175616836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003377748631564276,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21373271495103835,
      "num_tokens": 95021299.0,
      "step": 45785
    },
    {
      "entropy": 5.434230899810791,
      "epoch": 4.15366473149492,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033774379091969583,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21768580973148347,
      "num_tokens": 95031520.0,
      "step": 45790
    },
    {
      "entropy": 5.40581955909729,
      "epoch": 4.154118287373004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033771271738540634,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2235345795750618,
      "num_tokens": 95042127.0,
      "step": 45795
    },
    {
      "entropy": 5.490575361251831,
      "epoch": 4.154571843251088,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033768164255420184,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21472739577293395,
      "num_tokens": 95052181.0,
      "step": 45800
    },
    {
      "entropy": 5.520481014251709,
      "epoch": 4.155025399129173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033765056642672507,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21519790291786195,
      "num_tokens": 95062583.0,
      "step": 45805
    },
    {
      "entropy": 5.501868724822998,
      "epoch": 4.155478955007257,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033761948900361857,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21378954946994783,
      "num_tokens": 95073250.0,
      "step": 45810
    },
    {
      "entropy": 5.458225917816162,
      "epoch": 4.155932510885341,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003375884102855251,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21447069942951202,
      "num_tokens": 95083544.0,
      "step": 45815
    },
    {
      "entropy": 5.42863097190857,
      "epoch": 4.156386066763425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033755733027308746,
      "loss": 4.925,
      "mean_token_accuracy": 0.22200335562229156,
      "num_tokens": 95093933.0,
      "step": 45820
    },
    {
      "entropy": 5.452903842926025,
      "epoch": 4.15683962264151,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003375262489669481,
      "loss": 4.9944,
      "mean_token_accuracy": 0.21319639235734938,
      "num_tokens": 95104468.0,
      "step": 45825
    },
    {
      "entropy": 5.547179794311523,
      "epoch": 4.157293178519594,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033749516636775014,
      "loss": 4.984,
      "mean_token_accuracy": 0.2122375875711441,
      "num_tokens": 95114285.0,
      "step": 45830
    },
    {
      "entropy": 5.541571855545044,
      "epoch": 4.157746734397678,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033746408247613616,
      "loss": 4.9981,
      "mean_token_accuracy": 0.21319691240787506,
      "num_tokens": 95123984.0,
      "step": 45835
    },
    {
      "entropy": 5.537780141830444,
      "epoch": 4.1582002902757615,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000337432997292749,
      "loss": 5.0237,
      "mean_token_accuracy": 0.21245082318782807,
      "num_tokens": 95135466.0,
      "step": 45840
    },
    {
      "entropy": 5.405872869491577,
      "epoch": 4.158653846153846,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003374019108182315,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22107551991939545,
      "num_tokens": 95145223.0,
      "step": 45845
    },
    {
      "entropy": 5.493167591094971,
      "epoch": 4.15910740203193,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003373708230532267,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21474176943302153,
      "num_tokens": 95154916.0,
      "step": 45850
    },
    {
      "entropy": 5.449423789978027,
      "epoch": 4.159560957910014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033733973399837735,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21778870224952698,
      "num_tokens": 95164717.0,
      "step": 45855
    },
    {
      "entropy": 5.54533634185791,
      "epoch": 4.160014513788099,
      "grad_norm": 1.0,
      "learning_rate": 0.0003373086436543263,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2089369401335716,
      "num_tokens": 95175544.0,
      "step": 45860
    },
    {
      "entropy": 5.437351369857788,
      "epoch": 4.160468069666183,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003372775520217168,
      "loss": 5.0038,
      "mean_token_accuracy": 0.21205362379550935,
      "num_tokens": 95185997.0,
      "step": 45865
    },
    {
      "entropy": 5.438467121124267,
      "epoch": 4.160921625544267,
      "grad_norm": 1.0,
      "learning_rate": 0.0003372464591011916,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2084882766008377,
      "num_tokens": 95196967.0,
      "step": 45870
    },
    {
      "entropy": 5.437047147750855,
      "epoch": 4.161375181422351,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033721536489339374,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22569214403629304,
      "num_tokens": 95206788.0,
      "step": 45875
    },
    {
      "entropy": 5.544041347503662,
      "epoch": 4.161828737300436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033718426939896627,
      "loss": 4.9911,
      "mean_token_accuracy": 0.2158496931195259,
      "num_tokens": 95217806.0,
      "step": 45880
    },
    {
      "entropy": 5.470245122909546,
      "epoch": 4.16228229317852,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033715317261855235,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22559116035699844,
      "num_tokens": 95228311.0,
      "step": 45885
    },
    {
      "entropy": 5.404594850540161,
      "epoch": 4.162735849056604,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033712207455279485,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2165465161204338,
      "num_tokens": 95238872.0,
      "step": 45890
    },
    {
      "entropy": 5.355692434310913,
      "epoch": 4.163189404934688,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003370909752023371,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22473846673965453,
      "num_tokens": 95248575.0,
      "step": 45895
    },
    {
      "entropy": 5.466194152832031,
      "epoch": 4.163642960812772,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003370598745678221,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2098580151796341,
      "num_tokens": 95259323.0,
      "step": 45900
    },
    {
      "entropy": 5.498477602005005,
      "epoch": 4.164096516690856,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003370287726498932,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21653533577919007,
      "num_tokens": 95270018.0,
      "step": 45905
    },
    {
      "entropy": 5.430960035324096,
      "epoch": 4.16455007256894,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033699766944919325,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21864742189645767,
      "num_tokens": 95279783.0,
      "step": 45910
    },
    {
      "entropy": 5.465477180480957,
      "epoch": 4.165003628447025,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033696656496636586,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21312185674905776,
      "num_tokens": 95291200.0,
      "step": 45915
    },
    {
      "entropy": 5.468946075439453,
      "epoch": 4.165457184325109,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033693545920205416,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20637577772140503,
      "num_tokens": 95301641.0,
      "step": 45920
    },
    {
      "entropy": 5.418181896209717,
      "epoch": 4.165910740203193,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033690435215690125,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22739750146865845,
      "num_tokens": 95312636.0,
      "step": 45925
    },
    {
      "entropy": 5.4368589401245115,
      "epoch": 4.1663642960812775,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003368732438315506,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22308130115270614,
      "num_tokens": 95322168.0,
      "step": 45930
    },
    {
      "entropy": 5.408670902252197,
      "epoch": 4.166817851959362,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00033684213422664554,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2192107394337654,
      "num_tokens": 95333162.0,
      "step": 45935
    },
    {
      "entropy": 5.417633962631226,
      "epoch": 4.167271407837446,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003368110233428294,
      "loss": 4.9082,
      "mean_token_accuracy": 0.2178429737687111,
      "num_tokens": 95343806.0,
      "step": 45940
    },
    {
      "entropy": 5.448655223846435,
      "epoch": 4.167724963715529,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00033677991118074554,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22520391643047333,
      "num_tokens": 95353906.0,
      "step": 45945
    },
    {
      "entropy": 5.504672002792359,
      "epoch": 4.168178519593614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003367487977410373,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2058831885457039,
      "num_tokens": 95363801.0,
      "step": 45950
    },
    {
      "entropy": 5.533983325958252,
      "epoch": 4.168632075471698,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033671768302434823,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21188681870698928,
      "num_tokens": 95375012.0,
      "step": 45955
    },
    {
      "entropy": 5.507435607910156,
      "epoch": 4.169085631349782,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033668656703132174,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21738359779119493,
      "num_tokens": 95386429.0,
      "step": 45960
    },
    {
      "entropy": 5.414387273788452,
      "epoch": 4.169539187227866,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003366554497626014,
      "loss": 4.9325,
      "mean_token_accuracy": 0.21606046408414842,
      "num_tokens": 95396534.0,
      "step": 45965
    },
    {
      "entropy": 5.457985591888428,
      "epoch": 4.169992743105951,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033662433121883057,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21186640560626985,
      "num_tokens": 95406621.0,
      "step": 45970
    },
    {
      "entropy": 5.498845767974854,
      "epoch": 4.170446298984035,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003365932114006528,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20954303741455077,
      "num_tokens": 95417587.0,
      "step": 45975
    },
    {
      "entropy": 5.38740177154541,
      "epoch": 4.170899854862119,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00033656209030871184,
      "loss": 4.8169,
      "mean_token_accuracy": 0.22873683124780655,
      "num_tokens": 95428288.0,
      "step": 45980
    },
    {
      "entropy": 5.443341016769409,
      "epoch": 4.1713534107402035,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033653096794365115,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2179122805595398,
      "num_tokens": 95438093.0,
      "step": 45985
    },
    {
      "entropy": 5.372408390045166,
      "epoch": 4.171806966618288,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033649984430611437,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22170959860086442,
      "num_tokens": 95448397.0,
      "step": 45990
    },
    {
      "entropy": 5.494986295700073,
      "epoch": 4.172260522496371,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033646871939674506,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20267636924982071,
      "num_tokens": 95459533.0,
      "step": 45995
    },
    {
      "entropy": 5.477096176147461,
      "epoch": 4.172714078374455,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000336437593216187,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21902862191200256,
      "num_tokens": 95469507.0,
      "step": 46000
    },
    {
      "entropy": 5.552524042129517,
      "epoch": 4.17316763425254,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003364064657650839,
      "loss": 5.0122,
      "mean_token_accuracy": 0.20740258544683457,
      "num_tokens": 95479742.0,
      "step": 46005
    },
    {
      "entropy": 5.402548551559448,
      "epoch": 4.173621190130624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033637533704407936,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2247973844408989,
      "num_tokens": 95489751.0,
      "step": 46010
    },
    {
      "entropy": 5.506304407119751,
      "epoch": 4.174074746008708,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003363442070538173,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20999257564544677,
      "num_tokens": 95500580.0,
      "step": 46015
    },
    {
      "entropy": 5.527331733703614,
      "epoch": 4.1745283018867925,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033631307579494137,
      "loss": 5.0362,
      "mean_token_accuracy": 0.21260668486356735,
      "num_tokens": 95511018.0,
      "step": 46020
    },
    {
      "entropy": 5.499016523361206,
      "epoch": 4.174981857764877,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033628194326809543,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21740461140871048,
      "num_tokens": 95521220.0,
      "step": 46025
    },
    {
      "entropy": 5.434929037094117,
      "epoch": 4.175435413642961,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033625080947392324,
      "loss": 4.9423,
      "mean_token_accuracy": 0.2109615609049797,
      "num_tokens": 95531426.0,
      "step": 46030
    },
    {
      "entropy": 5.500299644470215,
      "epoch": 4.175888969521045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003362196744130687,
      "loss": 4.9656,
      "mean_token_accuracy": 0.22468343675136565,
      "num_tokens": 95542698.0,
      "step": 46035
    },
    {
      "entropy": 5.497317218780518,
      "epoch": 4.1763425253991295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033618853808617573,
      "loss": 4.904,
      "mean_token_accuracy": 0.2237196147441864,
      "num_tokens": 95553545.0,
      "step": 46040
    },
    {
      "entropy": 5.525575399398804,
      "epoch": 4.176796081277214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003361574004938882,
      "loss": 5.0127,
      "mean_token_accuracy": 0.21251944601535797,
      "num_tokens": 95564452.0,
      "step": 46045
    },
    {
      "entropy": 5.476856803894043,
      "epoch": 4.177249637155297,
      "grad_norm": 0.953125,
      "learning_rate": 0.00033612626163685,
      "loss": 4.985,
      "mean_token_accuracy": 0.22039155960083007,
      "num_tokens": 95574756.0,
      "step": 46050
    },
    {
      "entropy": 5.55009822845459,
      "epoch": 4.1777031930333814,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00033609512151570517,
      "loss": 5.0508,
      "mean_token_accuracy": 0.2076620876789093,
      "num_tokens": 95585147.0,
      "step": 46055
    },
    {
      "entropy": 5.475289964675904,
      "epoch": 4.178156748911466,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033606398013109765,
      "loss": 5.0718,
      "mean_token_accuracy": 0.21253828555345536,
      "num_tokens": 95596025.0,
      "step": 46060
    },
    {
      "entropy": 5.471476221084595,
      "epoch": 4.17861030478955,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003360328374836715,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21073208302259444,
      "num_tokens": 95607191.0,
      "step": 46065
    },
    {
      "entropy": 5.554716777801514,
      "epoch": 4.179063860667634,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003360016935740706,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2153700351715088,
      "num_tokens": 95617256.0,
      "step": 46070
    },
    {
      "entropy": 5.5448441982269285,
      "epoch": 4.1795174165457185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033597054840293927,
      "loss": 5.0033,
      "mean_token_accuracy": 0.21242849081754683,
      "num_tokens": 95627638.0,
      "step": 46075
    },
    {
      "entropy": 5.450800275802612,
      "epoch": 4.179970972423803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003359394019709214,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2145353302359581,
      "num_tokens": 95637662.0,
      "step": 46080
    },
    {
      "entropy": 5.445656633377075,
      "epoch": 4.180424528301887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033590825427866117,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2109634980559349,
      "num_tokens": 95647505.0,
      "step": 46085
    },
    {
      "entropy": 5.4730218887329105,
      "epoch": 4.180878084179971,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003358771053268027,
      "loss": 4.874,
      "mean_token_accuracy": 0.21166808903217316,
      "num_tokens": 95657810.0,
      "step": 46090
    },
    {
      "entropy": 5.409386539459229,
      "epoch": 4.181331640058056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033584595511599024,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21648083329200746,
      "num_tokens": 95667838.0,
      "step": 46095
    },
    {
      "entropy": 5.448362207412719,
      "epoch": 4.181785195936139,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003358148036468679,
      "loss": 4.9312,
      "mean_token_accuracy": 0.22229139804840087,
      "num_tokens": 95678497.0,
      "step": 46100
    },
    {
      "entropy": 5.380306911468506,
      "epoch": 4.182238751814223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033578365092007986,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2189726322889328,
      "num_tokens": 95688656.0,
      "step": 46105
    },
    {
      "entropy": 5.488941335678101,
      "epoch": 4.1826923076923075,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003357524969362705,
      "loss": 5.0364,
      "mean_token_accuracy": 0.2099490985274315,
      "num_tokens": 95699156.0,
      "step": 46110
    },
    {
      "entropy": 5.504911470413208,
      "epoch": 4.183145863570392,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033572134169608395,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21834742426872253,
      "num_tokens": 95709680.0,
      "step": 46115
    },
    {
      "entropy": 5.546048784255982,
      "epoch": 4.183599419448476,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033569018520016465,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21174290776252747,
      "num_tokens": 95719142.0,
      "step": 46120
    },
    {
      "entropy": 5.483978939056397,
      "epoch": 4.18405297532656,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003356590274491568,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20955527424812317,
      "num_tokens": 95729415.0,
      "step": 46125
    },
    {
      "entropy": 5.4201340675354,
      "epoch": 4.1845065312046446,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003356278684437049,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21196107119321822,
      "num_tokens": 95739775.0,
      "step": 46130
    },
    {
      "entropy": 5.52088680267334,
      "epoch": 4.184960087082729,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033559670818445315,
      "loss": 5.0153,
      "mean_token_accuracy": 0.21228066831827164,
      "num_tokens": 95751540.0,
      "step": 46135
    },
    {
      "entropy": 5.6380510330200195,
      "epoch": 4.185413642960813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000335565546672046,
      "loss": 5.1349,
      "mean_token_accuracy": 0.20883272886276244,
      "num_tokens": 95761885.0,
      "step": 46140
    },
    {
      "entropy": 5.57524938583374,
      "epoch": 4.185867198838897,
      "grad_norm": 1.09375,
      "learning_rate": 0.000335534383907128,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21560599654912949,
      "num_tokens": 95772284.0,
      "step": 46145
    },
    {
      "entropy": 5.4093180179595945,
      "epoch": 4.186320754716981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003355032198903434,
      "loss": 4.8967,
      "mean_token_accuracy": 0.21610979437828065,
      "num_tokens": 95783373.0,
      "step": 46150
    },
    {
      "entropy": 5.368869686126709,
      "epoch": 4.186774310595065,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033547205462233694,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22351455241441726,
      "num_tokens": 95793226.0,
      "step": 46155
    },
    {
      "entropy": 5.348373031616211,
      "epoch": 4.187227866473149,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00033544088810375287,
      "loss": 4.8511,
      "mean_token_accuracy": 0.21848566979169845,
      "num_tokens": 95803559.0,
      "step": 46160
    },
    {
      "entropy": 5.449258279800415,
      "epoch": 4.1876814223512335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003354097203352358,
      "loss": 4.9296,
      "mean_token_accuracy": 0.20999845266342163,
      "num_tokens": 95814620.0,
      "step": 46165
    },
    {
      "entropy": 5.430120897293091,
      "epoch": 4.188134978229318,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003353785513174305,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2207835629582405,
      "num_tokens": 95825024.0,
      "step": 46170
    },
    {
      "entropy": 5.512647438049316,
      "epoch": 4.188588534107402,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033534738105098124,
      "loss": 5.0102,
      "mean_token_accuracy": 0.21545128226280214,
      "num_tokens": 95837373.0,
      "step": 46175
    },
    {
      "entropy": 5.500385284423828,
      "epoch": 4.189042089985486,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033531620953653276,
      "loss": 4.942,
      "mean_token_accuracy": 0.213251756131649,
      "num_tokens": 95846962.0,
      "step": 46180
    },
    {
      "entropy": 5.445620441436768,
      "epoch": 4.189495645863571,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003352850367747297,
      "loss": 4.9772,
      "mean_token_accuracy": 0.2140626862645149,
      "num_tokens": 95857492.0,
      "step": 46185
    },
    {
      "entropy": 5.406223583221435,
      "epoch": 4.189949201741655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003352538627662168,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21674698889255523,
      "num_tokens": 95867903.0,
      "step": 46190
    },
    {
      "entropy": 5.556551361083985,
      "epoch": 4.190402757619739,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003352226875116386,
      "loss": 5.0782,
      "mean_token_accuracy": 0.21026356071233748,
      "num_tokens": 95878720.0,
      "step": 46195
    },
    {
      "entropy": 5.499082946777344,
      "epoch": 4.1908563134978225,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003351915110116399,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2062524676322937,
      "num_tokens": 95888406.0,
      "step": 46200
    },
    {
      "entropy": 5.465057611465454,
      "epoch": 4.191309869375907,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033516033326686537,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20417140424251556,
      "num_tokens": 95898431.0,
      "step": 46205
    },
    {
      "entropy": 5.484016847610474,
      "epoch": 4.191763425253991,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033512915427795975,
      "loss": 4.955,
      "mean_token_accuracy": 0.21915594339370728,
      "num_tokens": 95909089.0,
      "step": 46210
    },
    {
      "entropy": 5.422404098510742,
      "epoch": 4.192216981132075,
      "grad_norm": 1.34375,
      "learning_rate": 0.000335097974045568,
      "loss": 4.9425,
      "mean_token_accuracy": 0.21394274085760118,
      "num_tokens": 95919787.0,
      "step": 46215
    },
    {
      "entropy": 5.519698715209961,
      "epoch": 4.19267053701016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033506679257033485,
      "loss": 5.0272,
      "mean_token_accuracy": 0.21195320039987564,
      "num_tokens": 95930620.0,
      "step": 46220
    },
    {
      "entropy": 5.480221843719482,
      "epoch": 4.193124092888244,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003350356098529051,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2150789424777031,
      "num_tokens": 95940258.0,
      "step": 46225
    },
    {
      "entropy": 5.4892631530761715,
      "epoch": 4.193577648766328,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003350044258939236,
      "loss": 5.0098,
      "mean_token_accuracy": 0.21243503093719482,
      "num_tokens": 95950062.0,
      "step": 46230
    },
    {
      "entropy": 5.439295721054077,
      "epoch": 4.194031204644412,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033497324069403536,
      "loss": 4.8602,
      "mean_token_accuracy": 0.2166218414902687,
      "num_tokens": 95960658.0,
      "step": 46235
    },
    {
      "entropy": 5.4286596298217775,
      "epoch": 4.194484760522497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003349420542538851,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2204156458377838,
      "num_tokens": 95970497.0,
      "step": 46240
    },
    {
      "entropy": 5.466046953201294,
      "epoch": 4.194938316400581,
      "grad_norm": 1.0,
      "learning_rate": 0.00033491086657411797,
      "loss": 5.0316,
      "mean_token_accuracy": 0.2098434865474701,
      "num_tokens": 95980405.0,
      "step": 46245
    },
    {
      "entropy": 5.538436317443848,
      "epoch": 4.195391872278664,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003348796776553788,
      "loss": 5.0662,
      "mean_token_accuracy": 0.20596214085817338,
      "num_tokens": 95991111.0,
      "step": 46250
    },
    {
      "entropy": 5.592886304855346,
      "epoch": 4.1958454281567485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003348484874983127,
      "loss": 5.0068,
      "mean_token_accuracy": 0.214661905169487,
      "num_tokens": 96001962.0,
      "step": 46255
    },
    {
      "entropy": 5.483796262741089,
      "epoch": 4.196298984034833,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003348172961035646,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21560374945402144,
      "num_tokens": 96011774.0,
      "step": 46260
    },
    {
      "entropy": 5.440936136245727,
      "epoch": 4.196752539912917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003347861034717795,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21189493536949158,
      "num_tokens": 96021915.0,
      "step": 46265
    },
    {
      "entropy": 5.430316495895386,
      "epoch": 4.197206095791001,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033475490960360266,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21803874969482423,
      "num_tokens": 96032851.0,
      "step": 46270
    },
    {
      "entropy": 5.4201007843017575,
      "epoch": 4.197659651669086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000334723714499679,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2224498614668846,
      "num_tokens": 96042630.0,
      "step": 46275
    },
    {
      "entropy": 5.430074787139892,
      "epoch": 4.19811320754717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033469251816065375,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22475240528583526,
      "num_tokens": 96052154.0,
      "step": 46280
    },
    {
      "entropy": 5.494578409194946,
      "epoch": 4.198566763425254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000334661320587172,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21465315222740172,
      "num_tokens": 96062301.0,
      "step": 46285
    },
    {
      "entropy": 5.489022064208984,
      "epoch": 4.199020319303338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000334630121779879,
      "loss": 5.022,
      "mean_token_accuracy": 0.211589452624321,
      "num_tokens": 96072792.0,
      "step": 46290
    },
    {
      "entropy": 5.527511119842529,
      "epoch": 4.199473875181423,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033459892173941973,
      "loss": 5.0117,
      "mean_token_accuracy": 0.21172320246696472,
      "num_tokens": 96082610.0,
      "step": 46295
    },
    {
      "entropy": 5.538737487792969,
      "epoch": 4.199927431059507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003345677204664397,
      "loss": 5.0525,
      "mean_token_accuracy": 0.21614888310432434,
      "num_tokens": 96093095.0,
      "step": 46300
    },
    {
      "entropy": 5.464364671707154,
      "epoch": 4.20038098693759,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033453651796158397,
      "loss": 4.9385,
      "mean_token_accuracy": 0.22195885181427003,
      "num_tokens": 96103218.0,
      "step": 46305
    },
    {
      "entropy": 5.447591257095337,
      "epoch": 4.200834542815675,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033450531422549786,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2233414977788925,
      "num_tokens": 96113937.0,
      "step": 46310
    },
    {
      "entropy": 5.552868986129761,
      "epoch": 4.201288098693759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033447410925882673,
      "loss": 5.0901,
      "mean_token_accuracy": 0.21001553386449814,
      "num_tokens": 96124378.0,
      "step": 46315
    },
    {
      "entropy": 5.386086416244507,
      "epoch": 4.201741654571843,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003344429030622159,
      "loss": 4.8289,
      "mean_token_accuracy": 0.22245913296937941,
      "num_tokens": 96133657.0,
      "step": 46320
    },
    {
      "entropy": 5.445758819580078,
      "epoch": 4.202195210449927,
      "grad_norm": 1.125,
      "learning_rate": 0.00033441169563631055,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2205573320388794,
      "num_tokens": 96143669.0,
      "step": 46325
    },
    {
      "entropy": 5.446629285812378,
      "epoch": 4.202648766328012,
      "grad_norm": 1.0,
      "learning_rate": 0.00033438048698175625,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21560238897800446,
      "num_tokens": 96154782.0,
      "step": 46330
    },
    {
      "entropy": 5.4854433059692385,
      "epoch": 4.203102322206096,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003343492770991984,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21471561789512633,
      "num_tokens": 96165128.0,
      "step": 46335
    },
    {
      "entropy": 5.453725147247314,
      "epoch": 4.20355587808418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003343180659892824,
      "loss": 4.9595,
      "mean_token_accuracy": 0.2152429848909378,
      "num_tokens": 96176909.0,
      "step": 46340
    },
    {
      "entropy": 5.473302984237671,
      "epoch": 4.2040094339622645,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003342868536526536,
      "loss": 4.9483,
      "mean_token_accuracy": 0.22023711055517198,
      "num_tokens": 96188287.0,
      "step": 46345
    },
    {
      "entropy": 5.407210159301758,
      "epoch": 4.204462989840349,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033425564008995764,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21758031100034714,
      "num_tokens": 96198852.0,
      "step": 46350
    },
    {
      "entropy": 5.417476654052734,
      "epoch": 4.204916545718432,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003342244253018399,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21296060383319854,
      "num_tokens": 96208111.0,
      "step": 46355
    },
    {
      "entropy": 5.459411478042602,
      "epoch": 4.205370101596516,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003341932092889459,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2200110748410225,
      "num_tokens": 96218925.0,
      "step": 46360
    },
    {
      "entropy": 5.515427207946777,
      "epoch": 4.205823657474601,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003341619920519213,
      "loss": 4.9919,
      "mean_token_accuracy": 0.21642736047506334,
      "num_tokens": 96228573.0,
      "step": 46365
    },
    {
      "entropy": 5.463261270523072,
      "epoch": 4.206277213352685,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033413077359141166,
      "loss": 4.9696,
      "mean_token_accuracy": 0.213578237593174,
      "num_tokens": 96239109.0,
      "step": 46370
    },
    {
      "entropy": 5.511035680770874,
      "epoch": 4.206730769230769,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003340995539080625,
      "loss": 4.9885,
      "mean_token_accuracy": 0.2108746662735939,
      "num_tokens": 96249286.0,
      "step": 46375
    },
    {
      "entropy": 5.43310284614563,
      "epoch": 4.207184325108853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033406833300251947,
      "loss": 4.927,
      "mean_token_accuracy": 0.2198566749691963,
      "num_tokens": 96259515.0,
      "step": 46380
    },
    {
      "entropy": 5.4600257396698,
      "epoch": 4.207637880986938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003340371108754284,
      "loss": 4.894,
      "mean_token_accuracy": 0.219140687584877,
      "num_tokens": 96269136.0,
      "step": 46385
    },
    {
      "entropy": 5.5118073463439945,
      "epoch": 4.208091436865022,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003340058875274347,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2171775847673416,
      "num_tokens": 96279792.0,
      "step": 46390
    },
    {
      "entropy": 5.474096059799194,
      "epoch": 4.208544992743106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003339746629591842,
      "loss": 4.9954,
      "mean_token_accuracy": 0.2157618969678879,
      "num_tokens": 96289606.0,
      "step": 46395
    },
    {
      "entropy": 5.426948738098145,
      "epoch": 4.2089985486211905,
      "grad_norm": 0.953125,
      "learning_rate": 0.00033394343717132275,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21293589025735854,
      "num_tokens": 96299903.0,
      "step": 46400
    },
    {
      "entropy": 5.547938966751099,
      "epoch": 4.209452104499274,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033391221016449595,
      "loss": 5.0407,
      "mean_token_accuracy": 0.20631431490182878,
      "num_tokens": 96310179.0,
      "step": 46405
    },
    {
      "entropy": 5.5248631000518795,
      "epoch": 4.209905660377358,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033388098193934965,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2115866422653198,
      "num_tokens": 96320948.0,
      "step": 46410
    },
    {
      "entropy": 5.458063220977783,
      "epoch": 4.210359216255442,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003338497524965296,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21539010107517242,
      "num_tokens": 96332046.0,
      "step": 46415
    },
    {
      "entropy": 5.4272243022918705,
      "epoch": 4.210812772133527,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033381852183668166,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22080831378698348,
      "num_tokens": 96342327.0,
      "step": 46420
    },
    {
      "entropy": 5.440925407409668,
      "epoch": 4.211266328011611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003337872899604517,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21383659690618514,
      "num_tokens": 96352048.0,
      "step": 46425
    },
    {
      "entropy": 5.451179552078247,
      "epoch": 4.211719883889695,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033375605686848566,
      "loss": 4.877,
      "mean_token_accuracy": 0.22245041280984879,
      "num_tokens": 96362917.0,
      "step": 46430
    },
    {
      "entropy": 5.461952352523804,
      "epoch": 4.2121734397677795,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003337248225614293,
      "loss": 4.9712,
      "mean_token_accuracy": 0.21313309520483018,
      "num_tokens": 96373261.0,
      "step": 46435
    },
    {
      "entropy": 5.496547746658325,
      "epoch": 4.212626995645864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003336935870399288,
      "loss": 4.9863,
      "mean_token_accuracy": 0.20311009436845778,
      "num_tokens": 96384279.0,
      "step": 46440
    },
    {
      "entropy": 5.5542652130126955,
      "epoch": 4.213080551523948,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003336623503046299,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2127873942255974,
      "num_tokens": 96394071.0,
      "step": 46445
    },
    {
      "entropy": 5.473883056640625,
      "epoch": 4.213534107402032,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033363111235617854,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21865656971931458,
      "num_tokens": 96404233.0,
      "step": 46450
    },
    {
      "entropy": 5.486330890655518,
      "epoch": 4.2139876632801165,
      "grad_norm": 1.25,
      "learning_rate": 0.00033359987319522095,
      "loss": 4.9729,
      "mean_token_accuracy": 0.2152389332652092,
      "num_tokens": 96415388.0,
      "step": 46455
    },
    {
      "entropy": 5.502430248260498,
      "epoch": 4.2144412191582,
      "grad_norm": 1.0,
      "learning_rate": 0.00033356863282240295,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2144766703248024,
      "num_tokens": 96425744.0,
      "step": 46460
    },
    {
      "entropy": 5.471399354934692,
      "epoch": 4.214894775036284,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003335373912383707,
      "loss": 4.9549,
      "mean_token_accuracy": 0.22317763566970825,
      "num_tokens": 96436340.0,
      "step": 46465
    },
    {
      "entropy": 5.5514490604400635,
      "epoch": 4.2153483309143684,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003335061484437704,
      "loss": 5.0713,
      "mean_token_accuracy": 0.2072139099240303,
      "num_tokens": 96447384.0,
      "step": 46470
    },
    {
      "entropy": 5.473088121414184,
      "epoch": 4.215801886792453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000333474904439248,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21998565793037414,
      "num_tokens": 96456938.0,
      "step": 46475
    },
    {
      "entropy": 5.430836343765259,
      "epoch": 4.216255442670537,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003334436592254496,
      "loss": 4.966,
      "mean_token_accuracy": 0.21630011945962907,
      "num_tokens": 96467343.0,
      "step": 46480
    },
    {
      "entropy": 5.448326539993286,
      "epoch": 4.216708998548621,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033341241280302153,
      "loss": 4.9506,
      "mean_token_accuracy": 0.2167240098118782,
      "num_tokens": 96477443.0,
      "step": 46485
    },
    {
      "entropy": 5.456726360321045,
      "epoch": 4.2171625544267055,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033338116517260985,
      "loss": 4.9785,
      "mean_token_accuracy": 0.21475749611854553,
      "num_tokens": 96487711.0,
      "step": 46490
    },
    {
      "entropy": 5.608416652679443,
      "epoch": 4.21761611030479,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003333499163348608,
      "loss": 5.1047,
      "mean_token_accuracy": 0.20785392820835114,
      "num_tokens": 96497839.0,
      "step": 46495
    },
    {
      "entropy": 5.533189964294434,
      "epoch": 4.218069666182874,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033331866629042055,
      "loss": 5.0075,
      "mean_token_accuracy": 0.21500158160924912,
      "num_tokens": 96508237.0,
      "step": 46500
    },
    {
      "entropy": 5.43924880027771,
      "epoch": 4.218523222060958,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003332874150399355,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21437734812498094,
      "num_tokens": 96518877.0,
      "step": 46505
    },
    {
      "entropy": 5.449372291564941,
      "epoch": 4.218976777939042,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003332561625840519,
      "loss": 4.9327,
      "mean_token_accuracy": 0.216127672791481,
      "num_tokens": 96530734.0,
      "step": 46510
    },
    {
      "entropy": 5.508116245269775,
      "epoch": 4.219430333817126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003332249089234159,
      "loss": 5.0208,
      "mean_token_accuracy": 0.21589761674404145,
      "num_tokens": 96540983.0,
      "step": 46515
    },
    {
      "entropy": 5.524882221221924,
      "epoch": 4.21988388969521,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033319365405867394,
      "loss": 5.0486,
      "mean_token_accuracy": 0.20941655337810516,
      "num_tokens": 96551866.0,
      "step": 46520
    },
    {
      "entropy": 5.500959205627441,
      "epoch": 4.2203374455732945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033316239799047247,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21409401595592498,
      "num_tokens": 96562515.0,
      "step": 46525
    },
    {
      "entropy": 5.411762475967407,
      "epoch": 4.220791001451379,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033313114071945764,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21792432367801667,
      "num_tokens": 96573266.0,
      "step": 46530
    },
    {
      "entropy": 5.442257022857666,
      "epoch": 4.221244557329463,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003330998822462762,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22684218883514404,
      "num_tokens": 96583070.0,
      "step": 46535
    },
    {
      "entropy": 5.436363315582275,
      "epoch": 4.221698113207547,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033306862257157427,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21776268035173416,
      "num_tokens": 96593582.0,
      "step": 46540
    },
    {
      "entropy": 5.461607980728149,
      "epoch": 4.2221516690856316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003330373616959984,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22760380804538727,
      "num_tokens": 96603233.0,
      "step": 46545
    },
    {
      "entropy": 5.459699058532715,
      "epoch": 4.222605224963716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033300609962019523,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21655067056417465,
      "num_tokens": 96614361.0,
      "step": 46550
    },
    {
      "entropy": 5.4883417129516605,
      "epoch": 4.2230587808418,
      "grad_norm": 1.09375,
      "learning_rate": 0.000332974836344811,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21445755362510682,
      "num_tokens": 96624161.0,
      "step": 46555
    },
    {
      "entropy": 5.4664548397064205,
      "epoch": 4.2235123367198835,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033294357187049246,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21062770783901213,
      "num_tokens": 96634329.0,
      "step": 46560
    },
    {
      "entropy": 5.542921161651611,
      "epoch": 4.223965892597968,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033291230619788604,
      "loss": 5.0344,
      "mean_token_accuracy": 0.21344246566295624,
      "num_tokens": 96645575.0,
      "step": 46565
    },
    {
      "entropy": 5.485658884048462,
      "epoch": 4.224419448476052,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033288103932763846,
      "loss": 4.9584,
      "mean_token_accuracy": 0.2091584473848343,
      "num_tokens": 96656224.0,
      "step": 46570
    },
    {
      "entropy": 5.489381551742554,
      "epoch": 4.224873004354136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003328497712603962,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22509186267852782,
      "num_tokens": 96665950.0,
      "step": 46575
    },
    {
      "entropy": 5.375182104110718,
      "epoch": 4.2253265602322205,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00033281850199680586,
      "loss": 4.8301,
      "mean_token_accuracy": 0.22487298846244813,
      "num_tokens": 96676916.0,
      "step": 46580
    },
    {
      "entropy": 5.467363739013672,
      "epoch": 4.225780116110305,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003327872315375142,
      "loss": 5.047,
      "mean_token_accuracy": 0.20670240372419357,
      "num_tokens": 96687289.0,
      "step": 46585
    },
    {
      "entropy": 5.456803178787231,
      "epoch": 4.226233671988389,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003327559598831679,
      "loss": 4.9736,
      "mean_token_accuracy": 0.22005632668733596,
      "num_tokens": 96698149.0,
      "step": 46590
    },
    {
      "entropy": 5.446561288833618,
      "epoch": 4.226687227866473,
      "grad_norm": 1.0,
      "learning_rate": 0.00033272468703441363,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2174982264637947,
      "num_tokens": 96709466.0,
      "step": 46595
    },
    {
      "entropy": 5.5066211223602295,
      "epoch": 4.227140783744558,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000332693412991898,
      "loss": 4.9837,
      "mean_token_accuracy": 0.2145156279206276,
      "num_tokens": 96720179.0,
      "step": 46600
    },
    {
      "entropy": 5.537894439697266,
      "epoch": 4.227594339622642,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003326621377562681,
      "loss": 5.0087,
      "mean_token_accuracy": 0.21055252850055695,
      "num_tokens": 96730783.0,
      "step": 46605
    },
    {
      "entropy": 5.5222272872924805,
      "epoch": 4.228047895500726,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033263086132817035,
      "loss": 5.0535,
      "mean_token_accuracy": 0.2092081919312477,
      "num_tokens": 96740903.0,
      "step": 46610
    },
    {
      "entropy": 5.447844648361206,
      "epoch": 4.2285014513788095,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00033259958370825165,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22191937565803527,
      "num_tokens": 96750600.0,
      "step": 46615
    },
    {
      "entropy": 5.419514846801758,
      "epoch": 4.228955007256894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033256830489715905,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2170793056488037,
      "num_tokens": 96760550.0,
      "step": 46620
    },
    {
      "entropy": 5.476620864868164,
      "epoch": 4.229408563134978,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003325370248955392,
      "loss": 4.928,
      "mean_token_accuracy": 0.2194793000817299,
      "num_tokens": 96770784.0,
      "step": 46625
    },
    {
      "entropy": 5.432914304733276,
      "epoch": 4.229862119013062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003325057437040389,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22006072700023652,
      "num_tokens": 96780572.0,
      "step": 46630
    },
    {
      "entropy": 5.401108121871948,
      "epoch": 4.230315674891147,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003324744613233052,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21777316480875014,
      "num_tokens": 96790504.0,
      "step": 46635
    },
    {
      "entropy": 5.457557010650635,
      "epoch": 4.230769230769231,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033244317775398506,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21570458710193635,
      "num_tokens": 96800856.0,
      "step": 46640
    },
    {
      "entropy": 5.432721185684204,
      "epoch": 4.231222786647315,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003324118929967253,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21542917042970658,
      "num_tokens": 96811293.0,
      "step": 46645
    },
    {
      "entropy": 5.379464054107666,
      "epoch": 4.231676342525399,
      "grad_norm": 1.140625,
      "learning_rate": 0.000332380607052173,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2270624592900276,
      "num_tokens": 96821844.0,
      "step": 46650
    },
    {
      "entropy": 5.478475618362427,
      "epoch": 4.232129898403484,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003323493199209751,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20995557457208633,
      "num_tokens": 96833438.0,
      "step": 46655
    },
    {
      "entropy": 5.389040279388428,
      "epoch": 4.232583454281568,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003323180316037787,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21783771961927414,
      "num_tokens": 96843086.0,
      "step": 46660
    },
    {
      "entropy": 5.45471887588501,
      "epoch": 4.233037010159651,
      "grad_norm": 1.0,
      "learning_rate": 0.0003322867421012308,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22044066339731216,
      "num_tokens": 96853064.0,
      "step": 46665
    },
    {
      "entropy": 5.554422807693482,
      "epoch": 4.2334905660377355,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003322554514139784,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21233465671539306,
      "num_tokens": 96864213.0,
      "step": 46670
    },
    {
      "entropy": 5.603214693069458,
      "epoch": 4.23394412191582,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033222415954266873,
      "loss": 5.0844,
      "mean_token_accuracy": 0.2030336320400238,
      "num_tokens": 96875344.0,
      "step": 46675
    },
    {
      "entropy": 5.502106809616089,
      "epoch": 4.234397677793904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033219286648794886,
      "loss": 5.0141,
      "mean_token_accuracy": 0.21963839679956437,
      "num_tokens": 96885985.0,
      "step": 46680
    },
    {
      "entropy": 5.426137399673462,
      "epoch": 4.234851233671988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003321615722504659,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2255641758441925,
      "num_tokens": 96897512.0,
      "step": 46685
    },
    {
      "entropy": 5.452339601516724,
      "epoch": 4.235304789550073,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003321302768308671,
      "loss": 5.0176,
      "mean_token_accuracy": 0.210422283411026,
      "num_tokens": 96907630.0,
      "step": 46690
    },
    {
      "entropy": 5.403847408294678,
      "epoch": 4.235758345428157,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033209898022979964,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22047019600868226,
      "num_tokens": 96917400.0,
      "step": 46695
    },
    {
      "entropy": 5.559263849258423,
      "epoch": 4.236211901306241,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003320676824479106,
      "loss": 5.0628,
      "mean_token_accuracy": 0.20801408141851424,
      "num_tokens": 96927700.0,
      "step": 46700
    },
    {
      "entropy": 5.540345859527588,
      "epoch": 4.236665457184325,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003320363834858474,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21725286543369293,
      "num_tokens": 96938925.0,
      "step": 46705
    },
    {
      "entropy": 5.4068786144256595,
      "epoch": 4.23711901306241,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003320050833442573,
      "loss": 4.894,
      "mean_token_accuracy": 0.21584640592336654,
      "num_tokens": 96948956.0,
      "step": 46710
    },
    {
      "entropy": 5.406921291351319,
      "epoch": 4.237572568940493,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003319737820237875,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2164931356906891,
      "num_tokens": 96958565.0,
      "step": 46715
    },
    {
      "entropy": 5.439064979553223,
      "epoch": 4.238026124818577,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033194247952508544,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21283144503831863,
      "num_tokens": 96968261.0,
      "step": 46720
    },
    {
      "entropy": 5.524649715423584,
      "epoch": 4.238479680696662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033191117584879836,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21423934847116471,
      "num_tokens": 96978621.0,
      "step": 46725
    },
    {
      "entropy": 5.465154314041138,
      "epoch": 4.238933236574746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003318798709955736,
      "loss": 4.9179,
      "mean_token_accuracy": 0.21175342947244644,
      "num_tokens": 96989081.0,
      "step": 46730
    },
    {
      "entropy": 5.537933683395385,
      "epoch": 4.23938679245283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003318485649660586,
      "loss": 5.08,
      "mean_token_accuracy": 0.2073865219950676,
      "num_tokens": 97000518.0,
      "step": 46735
    },
    {
      "entropy": 5.4576843738555905,
      "epoch": 4.239840348330914,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033181725776090093,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21480202227830886,
      "num_tokens": 97012038.0,
      "step": 46740
    },
    {
      "entropy": 5.48263988494873,
      "epoch": 4.240293904208999,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033178594938074787,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21884433925151825,
      "num_tokens": 97022256.0,
      "step": 46745
    },
    {
      "entropy": 5.495461082458496,
      "epoch": 4.240747460087083,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003317546398262468,
      "loss": 5.0374,
      "mean_token_accuracy": 0.21017962098121643,
      "num_tokens": 97033063.0,
      "step": 46750
    },
    {
      "entropy": 5.43399338722229,
      "epoch": 4.241201015965167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003317233290980454,
      "loss": 5.0012,
      "mean_token_accuracy": 0.21336981654167175,
      "num_tokens": 97043711.0,
      "step": 46755
    },
    {
      "entropy": 5.496731519699097,
      "epoch": 4.2416545718432515,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033169201719679105,
      "loss": 4.914,
      "mean_token_accuracy": 0.2228263184428215,
      "num_tokens": 97053582.0,
      "step": 46760
    },
    {
      "entropy": 5.561380052566529,
      "epoch": 4.242108127721336,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00033166070412313133,
      "loss": 5.002,
      "mean_token_accuracy": 0.215406633913517,
      "num_tokens": 97064535.0,
      "step": 46765
    },
    {
      "entropy": 5.4674482345581055,
      "epoch": 4.242561683599419,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003316293898777138,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20400147438049315,
      "num_tokens": 97074295.0,
      "step": 46770
    },
    {
      "entropy": 5.447279548645019,
      "epoch": 4.243015239477503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033159807446118614,
      "loss": 4.9524,
      "mean_token_accuracy": 0.208495032787323,
      "num_tokens": 97084781.0,
      "step": 46775
    },
    {
      "entropy": 5.572683334350586,
      "epoch": 4.243468795355588,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003315667578741958,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20927851796150207,
      "num_tokens": 97094848.0,
      "step": 46780
    },
    {
      "entropy": 5.500085878372192,
      "epoch": 4.243922351233672,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003315354401173905,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2178793281316757,
      "num_tokens": 97104747.0,
      "step": 46785
    },
    {
      "entropy": 5.408771181106568,
      "epoch": 4.244375907111756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003315041211914179,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21509658843278884,
      "num_tokens": 97115599.0,
      "step": 46790
    },
    {
      "entropy": 5.478577136993408,
      "epoch": 4.24482946298984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003314728010969257,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2155631199479103,
      "num_tokens": 97127492.0,
      "step": 46795
    },
    {
      "entropy": 5.536603164672852,
      "epoch": 4.245283018867925,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033144147983456154,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2072212353348732,
      "num_tokens": 97138336.0,
      "step": 46800
    },
    {
      "entropy": 5.4806201457977295,
      "epoch": 4.245736574746009,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003314101574049732,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2204006090760231,
      "num_tokens": 97149196.0,
      "step": 46805
    },
    {
      "entropy": 5.49855785369873,
      "epoch": 4.246190130624093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033137883380880847,
      "loss": 4.9952,
      "mean_token_accuracy": 0.21416943967342378,
      "num_tokens": 97159622.0,
      "step": 46810
    },
    {
      "entropy": 5.54072937965393,
      "epoch": 4.2466436865021775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000331347509046715,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20718554109334947,
      "num_tokens": 97170149.0,
      "step": 46815
    },
    {
      "entropy": 5.525779294967651,
      "epoch": 4.247097242380261,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003313161831193408,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21061777770519258,
      "num_tokens": 97180409.0,
      "step": 46820
    },
    {
      "entropy": 5.5018867492675785,
      "epoch": 4.247550798258345,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033128485602733354,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21256036162376404,
      "num_tokens": 97190899.0,
      "step": 46825
    },
    {
      "entropy": 5.498130464553833,
      "epoch": 4.248004354136429,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003312535277713411,
      "loss": 4.9864,
      "mean_token_accuracy": 0.22121568769216537,
      "num_tokens": 97200759.0,
      "step": 46830
    },
    {
      "entropy": 5.4959746360778805,
      "epoch": 4.248457910014514,
      "grad_norm": 0.984375,
      "learning_rate": 0.00033122219835201135,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21801339089870453,
      "num_tokens": 97211476.0,
      "step": 46835
    },
    {
      "entropy": 5.550125598907471,
      "epoch": 4.248911465892598,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003311908677699923,
      "loss": 5.037,
      "mean_token_accuracy": 0.20922426879405975,
      "num_tokens": 97222374.0,
      "step": 46840
    },
    {
      "entropy": 5.505874824523926,
      "epoch": 4.249365021770682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003311595360259317,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21756287068128585,
      "num_tokens": 97233079.0,
      "step": 46845
    },
    {
      "entropy": 5.393514347076416,
      "epoch": 4.2498185776487665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033112820312047756,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22107508331537246,
      "num_tokens": 97243796.0,
      "step": 46850
    },
    {
      "entropy": 5.431392097473145,
      "epoch": 4.250272133526851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033109686905427794,
      "loss": 4.8422,
      "mean_token_accuracy": 0.22626696527004242,
      "num_tokens": 97253341.0,
      "step": 46855
    },
    {
      "entropy": 5.5340831756591795,
      "epoch": 4.250725689404935,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00033106553382798075,
      "loss": 5.0158,
      "mean_token_accuracy": 0.21011691838502883,
      "num_tokens": 97264244.0,
      "step": 46860
    },
    {
      "entropy": 5.432056713104248,
      "epoch": 4.251179245283019,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033103419744223413,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22037981003522872,
      "num_tokens": 97274239.0,
      "step": 46865
    },
    {
      "entropy": 5.4082605838775635,
      "epoch": 4.251632801161103,
      "grad_norm": 1.0,
      "learning_rate": 0.0003310028598976859,
      "loss": 4.831,
      "mean_token_accuracy": 0.22783569246530533,
      "num_tokens": 97284709.0,
      "step": 46870
    },
    {
      "entropy": 5.440661668777466,
      "epoch": 4.252086357039187,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00033097152119498425,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2165185406804085,
      "num_tokens": 97295262.0,
      "step": 46875
    },
    {
      "entropy": 5.4758062839508055,
      "epoch": 4.252539912917271,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003309401813347774,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21171786934137343,
      "num_tokens": 97306143.0,
      "step": 46880
    },
    {
      "entropy": 5.4886260509490965,
      "epoch": 4.2529934687953554,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003309088403177132,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22392154037952422,
      "num_tokens": 97316248.0,
      "step": 46885
    },
    {
      "entropy": 5.521383047103882,
      "epoch": 4.25344702467344,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033087749814444,
      "loss": 5.0528,
      "mean_token_accuracy": 0.21049629747867585,
      "num_tokens": 97327704.0,
      "step": 46890
    },
    {
      "entropy": 5.546318006515503,
      "epoch": 4.253900580551524,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033084615481560585,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2056211471557617,
      "num_tokens": 97337817.0,
      "step": 46895
    },
    {
      "entropy": 5.5416646003723145,
      "epoch": 4.254354136429608,
      "grad_norm": 1.0,
      "learning_rate": 0.00033081481033185903,
      "loss": 5.0384,
      "mean_token_accuracy": 0.21175303012132646,
      "num_tokens": 97348315.0,
      "step": 46900
    },
    {
      "entropy": 5.458838987350464,
      "epoch": 4.2548076923076925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003307834646938477,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21435348987579345,
      "num_tokens": 97358842.0,
      "step": 46905
    },
    {
      "entropy": 5.545959997177124,
      "epoch": 4.255261248185777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003307521179022201,
      "loss": 5.0632,
      "mean_token_accuracy": 0.21030620336532593,
      "num_tokens": 97369247.0,
      "step": 46910
    },
    {
      "entropy": 5.495222187042236,
      "epoch": 4.255714804063861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003307207699576246,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21921750903129578,
      "num_tokens": 97379668.0,
      "step": 46915
    },
    {
      "entropy": 5.473016452789307,
      "epoch": 4.256168359941945,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033068942086070924,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2180556297302246,
      "num_tokens": 97389131.0,
      "step": 46920
    },
    {
      "entropy": 5.494143056869507,
      "epoch": 4.256621915820029,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033065807061212254,
      "loss": 5.0692,
      "mean_token_accuracy": 0.20634850710630417,
      "num_tokens": 97400583.0,
      "step": 46925
    },
    {
      "entropy": 5.524472427368164,
      "epoch": 4.257075471698113,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003306267192125128,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21495823562145233,
      "num_tokens": 97410828.0,
      "step": 46930
    },
    {
      "entropy": 5.513527536392212,
      "epoch": 4.257529027576197,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003305953666625283,
      "loss": 5.004,
      "mean_token_accuracy": 0.21016470491886138,
      "num_tokens": 97421190.0,
      "step": 46935
    },
    {
      "entropy": 5.486300945281982,
      "epoch": 4.2579825834542815,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00033056401296281733,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20798340439796448,
      "num_tokens": 97432257.0,
      "step": 46940
    },
    {
      "entropy": 5.469662523269653,
      "epoch": 4.258436139332366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003305326581140286,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2138917475938797,
      "num_tokens": 97442272.0,
      "step": 46945
    },
    {
      "entropy": 5.4724830150604244,
      "epoch": 4.25888969521045,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033050130211681017,
      "loss": 4.9484,
      "mean_token_accuracy": 0.22293840795755387,
      "num_tokens": 97451462.0,
      "step": 46950
    },
    {
      "entropy": 5.441675615310669,
      "epoch": 4.259343251088534,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033046994497181077,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2186179831624031,
      "num_tokens": 97461579.0,
      "step": 46955
    },
    {
      "entropy": 5.564209842681885,
      "epoch": 4.2597968069666186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033043858667967875,
      "loss": 5.0501,
      "mean_token_accuracy": 0.21262132823467256,
      "num_tokens": 97472892.0,
      "step": 46960
    },
    {
      "entropy": 5.451684045791626,
      "epoch": 4.260250362844703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033040722724106257,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21265192925930024,
      "num_tokens": 97484278.0,
      "step": 46965
    },
    {
      "entropy": 5.445078563690186,
      "epoch": 4.260703918722786,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00033037586665661094,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21227608621120453,
      "num_tokens": 97494127.0,
      "step": 46970
    },
    {
      "entropy": 5.446001577377319,
      "epoch": 4.2611574746008705,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003303445049269722,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21596609950065612,
      "num_tokens": 97504213.0,
      "step": 46975
    },
    {
      "entropy": 5.5588373184204105,
      "epoch": 4.261611030478955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033031314205279503,
      "loss": 5.0386,
      "mean_token_accuracy": 0.212447389960289,
      "num_tokens": 97515242.0,
      "step": 46980
    },
    {
      "entropy": 5.543279027938842,
      "epoch": 4.262064586357039,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033028177803472793,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21829483807086944,
      "num_tokens": 97525956.0,
      "step": 46985
    },
    {
      "entropy": 5.399789142608642,
      "epoch": 4.262518142235123,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003302504128734196,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22372492253780366,
      "num_tokens": 97535973.0,
      "step": 46990
    },
    {
      "entropy": 5.441490697860718,
      "epoch": 4.2629716981132075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033021904656951867,
      "loss": 4.9398,
      "mean_token_accuracy": 0.22431570887565613,
      "num_tokens": 97547580.0,
      "step": 46995
    },
    {
      "entropy": 5.505146265029907,
      "epoch": 4.263425253991292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00033018767912367376,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21795428842306136,
      "num_tokens": 97558050.0,
      "step": 47000
    },
    {
      "entropy": 5.511179256439209,
      "epoch": 4.263878809869376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033015631053653357,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2234828755259514,
      "num_tokens": 97567997.0,
      "step": 47005
    },
    {
      "entropy": 5.40337553024292,
      "epoch": 4.26433236574746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003301249408087469,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22055685371160508,
      "num_tokens": 97577220.0,
      "step": 47010
    },
    {
      "entropy": 5.427917337417602,
      "epoch": 4.264785921625545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033009356994096237,
      "loss": 4.9593,
      "mean_token_accuracy": 0.22037215232849122,
      "num_tokens": 97588421.0,
      "step": 47015
    },
    {
      "entropy": 5.370685386657715,
      "epoch": 4.265239477503629,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00033006219793382867,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21909355670213698,
      "num_tokens": 97598624.0,
      "step": 47020
    },
    {
      "entropy": 5.494549894332886,
      "epoch": 4.265693033381712,
      "grad_norm": 1.125,
      "learning_rate": 0.0003300308247879948,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21086453199386596,
      "num_tokens": 97609128.0,
      "step": 47025
    },
    {
      "entropy": 5.497999382019043,
      "epoch": 4.2661465892597965,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003299994505041094,
      "loss": 4.9677,
      "mean_token_accuracy": 0.2212685227394104,
      "num_tokens": 97619549.0,
      "step": 47030
    },
    {
      "entropy": 5.467164134979248,
      "epoch": 4.266600145137881,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00032996807508282134,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2143237978219986,
      "num_tokens": 97630184.0,
      "step": 47035
    },
    {
      "entropy": 5.471703243255615,
      "epoch": 4.267053701015965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032993669852477954,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22099502384662628,
      "num_tokens": 97640569.0,
      "step": 47040
    },
    {
      "entropy": 5.436417293548584,
      "epoch": 4.267507256894049,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003299053208306328,
      "loss": 4.9017,
      "mean_token_accuracy": 0.2187468960881233,
      "num_tokens": 97651006.0,
      "step": 47045
    },
    {
      "entropy": 5.441226482391357,
      "epoch": 4.267960812772134,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003298739420010299,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21697217375040054,
      "num_tokens": 97661567.0,
      "step": 47050
    },
    {
      "entropy": 5.479335308074951,
      "epoch": 4.268414368650218,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003298425620366201,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2206734076142311,
      "num_tokens": 97673058.0,
      "step": 47055
    },
    {
      "entropy": 5.613692283630371,
      "epoch": 4.268867924528302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003298111809380521,
      "loss": 5.08,
      "mean_token_accuracy": 0.205881205201149,
      "num_tokens": 97684509.0,
      "step": 47060
    },
    {
      "entropy": 5.546368646621704,
      "epoch": 4.269321480406386,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032977979870597477,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21427170634269715,
      "num_tokens": 97693954.0,
      "step": 47065
    },
    {
      "entropy": 5.465792655944824,
      "epoch": 4.269775036284471,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003297484153410374,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22061142921447754,
      "num_tokens": 97704345.0,
      "step": 47070
    },
    {
      "entropy": 5.341120195388794,
      "epoch": 4.270228592162554,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032971703084388885,
      "loss": 4.831,
      "mean_token_accuracy": 0.2341817781329155,
      "num_tokens": 97714857.0,
      "step": 47075
    },
    {
      "entropy": 5.435633039474487,
      "epoch": 4.270682148040638,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000329685645215178,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21575847268104553,
      "num_tokens": 97725921.0,
      "step": 47080
    },
    {
      "entropy": 5.444227075576782,
      "epoch": 4.2711357039187225,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032965425845555413,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22350773811340333,
      "num_tokens": 97735361.0,
      "step": 47085
    },
    {
      "entropy": 5.47302393913269,
      "epoch": 4.271589259796807,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032962287056566633,
      "loss": 4.9649,
      "mean_token_accuracy": 0.21055385023355483,
      "num_tokens": 97745915.0,
      "step": 47090
    },
    {
      "entropy": 5.507869529724121,
      "epoch": 4.272042815674891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003295914815461637,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2219053715467453,
      "num_tokens": 97755663.0,
      "step": 47095
    },
    {
      "entropy": 5.514576959609985,
      "epoch": 4.272496371552975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032956009139769514,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20614483058452607,
      "num_tokens": 97766300.0,
      "step": 47100
    },
    {
      "entropy": 5.510130500793457,
      "epoch": 4.27294992743106,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032952870012091013,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21727784872055053,
      "num_tokens": 97776584.0,
      "step": 47105
    },
    {
      "entropy": 5.434438753128052,
      "epoch": 4.273403483309144,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032949730771645767,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22215356081724166,
      "num_tokens": 97786230.0,
      "step": 47110
    },
    {
      "entropy": 5.486904430389404,
      "epoch": 4.273857039187228,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000329465914184987,
      "loss": 5.0103,
      "mean_token_accuracy": 0.21176085621118546,
      "num_tokens": 97796671.0,
      "step": 47115
    },
    {
      "entropy": 5.412958288192749,
      "epoch": 4.274310595065312,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003294345195271473,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2190636694431305,
      "num_tokens": 97807569.0,
      "step": 47120
    },
    {
      "entropy": 5.585106992721558,
      "epoch": 4.274764150943396,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032940312374358794,
      "loss": 5.1021,
      "mean_token_accuracy": 0.2078483983874321,
      "num_tokens": 97818680.0,
      "step": 47125
    },
    {
      "entropy": 5.514622497558594,
      "epoch": 4.27521770682148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000329371726834958,
      "loss": 4.9719,
      "mean_token_accuracy": 0.2156983032822609,
      "num_tokens": 97828955.0,
      "step": 47130
    },
    {
      "entropy": 5.463759517669677,
      "epoch": 4.275671262699564,
      "grad_norm": 1.015625,
      "learning_rate": 0.000329340328801907,
      "loss": 4.941,
      "mean_token_accuracy": 0.2250613033771515,
      "num_tokens": 97839000.0,
      "step": 47135
    },
    {
      "entropy": 5.452719068527221,
      "epoch": 4.276124818577649,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003293089296450841,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20725660920143127,
      "num_tokens": 97850140.0,
      "step": 47140
    },
    {
      "entropy": 5.472657203674316,
      "epoch": 4.276578374455733,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003292775293651388,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22089338600635527,
      "num_tokens": 97860887.0,
      "step": 47145
    },
    {
      "entropy": 5.490013837814331,
      "epoch": 4.277031930333817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003292461279627203,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21593645215034485,
      "num_tokens": 97871190.0,
      "step": 47150
    },
    {
      "entropy": 5.448474740982055,
      "epoch": 4.277485486211901,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032921472543847804,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22115544974803925,
      "num_tokens": 97881479.0,
      "step": 47155
    },
    {
      "entropy": 5.424573755264282,
      "epoch": 4.277939042089986,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003291833217930614,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21993397921323776,
      "num_tokens": 97892340.0,
      "step": 47160
    },
    {
      "entropy": 5.490475845336914,
      "epoch": 4.27839259796807,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032915191702711995,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2146371126174927,
      "num_tokens": 97902882.0,
      "step": 47165
    },
    {
      "entropy": 5.465791034698486,
      "epoch": 4.278846153846154,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032912051114130297,
      "loss": 4.8622,
      "mean_token_accuracy": 0.219920714199543,
      "num_tokens": 97913187.0,
      "step": 47170
    },
    {
      "entropy": 5.441436719894409,
      "epoch": 4.2792997097242385,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032908910413626,
      "loss": 4.8749,
      "mean_token_accuracy": 0.23035672456026077,
      "num_tokens": 97922434.0,
      "step": 47175
    },
    {
      "entropy": 5.327313184738159,
      "epoch": 4.279753265602322,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003290576960126407,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22294857800006868,
      "num_tokens": 97933020.0,
      "step": 47180
    },
    {
      "entropy": 5.374651384353638,
      "epoch": 4.280206821480406,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032902628677109434,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22020450681447984,
      "num_tokens": 97944103.0,
      "step": 47185
    },
    {
      "entropy": 5.504331159591675,
      "epoch": 4.28066037735849,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003289948764122707,
      "loss": 4.923,
      "mean_token_accuracy": 0.22352461516857147,
      "num_tokens": 97953692.0,
      "step": 47190
    },
    {
      "entropy": 5.502703857421875,
      "epoch": 4.281113933236575,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003289634649368193,
      "loss": 4.9504,
      "mean_token_accuracy": 0.22307925522327424,
      "num_tokens": 97963592.0,
      "step": 47195
    },
    {
      "entropy": 5.553171396255493,
      "epoch": 4.281567489114659,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003289320523453895,
      "loss": 5.1375,
      "mean_token_accuracy": 0.20731363892555238,
      "num_tokens": 97974092.0,
      "step": 47200
    },
    {
      "entropy": 5.416953992843628,
      "epoch": 4.282021044992743,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032890063863863126,
      "loss": 4.832,
      "mean_token_accuracy": 0.22166794091463088,
      "num_tokens": 97984175.0,
      "step": 47205
    },
    {
      "entropy": 5.421210908889771,
      "epoch": 4.282474600870827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003288692238171941,
      "loss": 4.9446,
      "mean_token_accuracy": 0.23288452476263047,
      "num_tokens": 97994993.0,
      "step": 47210
    },
    {
      "entropy": 5.480302143096924,
      "epoch": 4.282928156748912,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00032883780788172756,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20961487740278245,
      "num_tokens": 98005892.0,
      "step": 47215
    },
    {
      "entropy": 5.4421830654144285,
      "epoch": 4.283381712626996,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003288063908328814,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2183000013232231,
      "num_tokens": 98016726.0,
      "step": 47220
    },
    {
      "entropy": 5.492641401290894,
      "epoch": 4.28383526850508,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003287749726713055,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21456893980503083,
      "num_tokens": 98028500.0,
      "step": 47225
    },
    {
      "entropy": 5.547808742523193,
      "epoch": 4.284288824383164,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003287435533976494,
      "loss": 5.0606,
      "mean_token_accuracy": 0.20837472826242448,
      "num_tokens": 98039304.0,
      "step": 47230
    },
    {
      "entropy": 5.481419372558594,
      "epoch": 4.284742380261248,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003287121330125629,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21179343312978743,
      "num_tokens": 98049448.0,
      "step": 47235
    },
    {
      "entropy": 5.484037256240844,
      "epoch": 4.285195936139332,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032868071151669585,
      "loss": 4.9637,
      "mean_token_accuracy": 0.22327186912298203,
      "num_tokens": 98059532.0,
      "step": 47240
    },
    {
      "entropy": 5.424869632720947,
      "epoch": 4.285649492017416,
      "grad_norm": 1.140625,
      "learning_rate": 0.000328649288910698,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22467549741268159,
      "num_tokens": 98069385.0,
      "step": 47245
    },
    {
      "entropy": 5.486191129684448,
      "epoch": 4.286103047895501,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003286178651952192,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21691755801439286,
      "num_tokens": 98079354.0,
      "step": 47250
    },
    {
      "entropy": 5.512998914718628,
      "epoch": 4.286556603773585,
      "grad_norm": 0.96875,
      "learning_rate": 0.00032858644037090924,
      "loss": 5.0664,
      "mean_token_accuracy": 0.2080034539103508,
      "num_tokens": 98090864.0,
      "step": 47255
    },
    {
      "entropy": 5.495344829559326,
      "epoch": 4.287010159651669,
      "grad_norm": 2.234375,
      "learning_rate": 0.0003285550144384181,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21552201807498933,
      "num_tokens": 98101515.0,
      "step": 47260
    },
    {
      "entropy": 5.5276671886444095,
      "epoch": 4.2874637155297535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032852358739839555,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2132853165268898,
      "num_tokens": 98111786.0,
      "step": 47265
    },
    {
      "entropy": 5.438475799560547,
      "epoch": 4.287917271407838,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032849215925149155,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22637722939252852,
      "num_tokens": 98121732.0,
      "step": 47270
    },
    {
      "entropy": 5.496871471405029,
      "epoch": 4.288370827285922,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003284607299983561,
      "loss": 5.0594,
      "mean_token_accuracy": 0.20930132120847703,
      "num_tokens": 98131417.0,
      "step": 47275
    },
    {
      "entropy": 5.544427299499512,
      "epoch": 4.288824383164005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032842929963963915,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21580762267112732,
      "num_tokens": 98141664.0,
      "step": 47280
    },
    {
      "entropy": 5.474904346466064,
      "epoch": 4.28927793904209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032839786817599065,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21568366140127182,
      "num_tokens": 98150959.0,
      "step": 47285
    },
    {
      "entropy": 5.485706758499146,
      "epoch": 4.289731494920174,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003283664356080606,
      "loss": 5.0329,
      "mean_token_accuracy": 0.21378330737352372,
      "num_tokens": 98161957.0,
      "step": 47290
    },
    {
      "entropy": 5.470439672470093,
      "epoch": 4.290185050798258,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032833500193649916,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22177816182374954,
      "num_tokens": 98172053.0,
      "step": 47295
    },
    {
      "entropy": 5.502331209182739,
      "epoch": 4.2906386066763424,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032830356716195624,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21533995419740676,
      "num_tokens": 98182074.0,
      "step": 47300
    },
    {
      "entropy": 5.512929058074951,
      "epoch": 4.291092162554427,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003282721312850819,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2174337163567543,
      "num_tokens": 98192591.0,
      "step": 47305
    },
    {
      "entropy": 5.41845293045044,
      "epoch": 4.291545718432511,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003282406943065264,
      "loss": 4.8578,
      "mean_token_accuracy": 0.2231736361980438,
      "num_tokens": 98203426.0,
      "step": 47310
    },
    {
      "entropy": 5.4234823226928714,
      "epoch": 4.291999274310595,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032820925622693977,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21986007690429688,
      "num_tokens": 98214342.0,
      "step": 47315
    },
    {
      "entropy": 5.497733879089355,
      "epoch": 4.2924528301886795,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032817781704697204,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2116248279809952,
      "num_tokens": 98224155.0,
      "step": 47320
    },
    {
      "entropy": 5.486463451385498,
      "epoch": 4.292906386066764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003281463767672736,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21975968033075333,
      "num_tokens": 98233613.0,
      "step": 47325
    },
    {
      "entropy": 5.465371942520141,
      "epoch": 4.293359941944848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003281149353884946,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21950771957635878,
      "num_tokens": 98243117.0,
      "step": 47330
    },
    {
      "entropy": 5.422942304611206,
      "epoch": 4.293813497822931,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003280834929112851,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22176173329353333,
      "num_tokens": 98253619.0,
      "step": 47335
    },
    {
      "entropy": 5.435182571411133,
      "epoch": 4.294267053701016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003280520493362955,
      "loss": 4.9372,
      "mean_token_accuracy": 0.22086515575647353,
      "num_tokens": 98264221.0,
      "step": 47340
    },
    {
      "entropy": 5.488523626327515,
      "epoch": 4.2947206095791,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003280206046641759,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21789761781692504,
      "num_tokens": 98275530.0,
      "step": 47345
    },
    {
      "entropy": 5.585740089416504,
      "epoch": 4.295174165457184,
      "grad_norm": 0.953125,
      "learning_rate": 0.00032798915889557687,
      "loss": 5.0379,
      "mean_token_accuracy": 0.21021779030561447,
      "num_tokens": 98285937.0,
      "step": 47350
    },
    {
      "entropy": 5.483759355545044,
      "epoch": 4.2956277213352685,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003279577120311483,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2174907848238945,
      "num_tokens": 98295690.0,
      "step": 47355
    },
    {
      "entropy": 5.382821273803711,
      "epoch": 4.296081277213353,
      "grad_norm": 1.078125,
      "learning_rate": 0.000327926264071541,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21963433176279068,
      "num_tokens": 98305073.0,
      "step": 47360
    },
    {
      "entropy": 5.504685163497925,
      "epoch": 4.296534833091437,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003278948150174049,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20451211035251618,
      "num_tokens": 98316008.0,
      "step": 47365
    },
    {
      "entropy": 5.473961448669433,
      "epoch": 4.296988388969521,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003278633648693906,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21591684818267823,
      "num_tokens": 98327032.0,
      "step": 47370
    },
    {
      "entropy": 5.512147760391235,
      "epoch": 4.2974419448476056,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003278319136281484,
      "loss": 4.9719,
      "mean_token_accuracy": 0.22050662487745284,
      "num_tokens": 98336085.0,
      "step": 47375
    },
    {
      "entropy": 5.496565055847168,
      "epoch": 4.297895500725689,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032780046129432885,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2204699158668518,
      "num_tokens": 98346180.0,
      "step": 47380
    },
    {
      "entropy": 5.466293478012085,
      "epoch": 4.298349056603773,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032776900786858214,
      "loss": 4.8942,
      "mean_token_accuracy": 0.2178386226296425,
      "num_tokens": 98356001.0,
      "step": 47385
    },
    {
      "entropy": 5.49166841506958,
      "epoch": 4.2988026124818575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000327737553351559,
      "loss": 5.0255,
      "mean_token_accuracy": 0.21416840702295303,
      "num_tokens": 98366744.0,
      "step": 47390
    },
    {
      "entropy": 5.3904969692230225,
      "epoch": 4.299256168359942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032770609774390985,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22153188437223434,
      "num_tokens": 98377350.0,
      "step": 47395
    },
    {
      "entropy": 5.4408070087432865,
      "epoch": 4.299709724238026,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003276746410462852,
      "loss": 4.9163,
      "mean_token_accuracy": 0.22398261576890946,
      "num_tokens": 98387926.0,
      "step": 47400
    },
    {
      "entropy": 5.539003849029541,
      "epoch": 4.30016328011611,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032764318325933546,
      "loss": 5.0008,
      "mean_token_accuracy": 0.21687127500772477,
      "num_tokens": 98397117.0,
      "step": 47405
    },
    {
      "entropy": 5.545068359375,
      "epoch": 4.3006168359941945,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003276117243837113,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21261948347091675,
      "num_tokens": 98407983.0,
      "step": 47410
    },
    {
      "entropy": 5.470067882537842,
      "epoch": 4.301070391872279,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003275802644200633,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21007515788078307,
      "num_tokens": 98419940.0,
      "step": 47415
    },
    {
      "entropy": 5.42319769859314,
      "epoch": 4.301523947750363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000327548803369042,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22298619151115417,
      "num_tokens": 98430487.0,
      "step": 47420
    },
    {
      "entropy": 5.479446840286255,
      "epoch": 4.301977503628447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032751734123129814,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22387623339891433,
      "num_tokens": 98440231.0,
      "step": 47425
    },
    {
      "entropy": 5.510072374343872,
      "epoch": 4.302431059506532,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003274858780074822,
      "loss": 5.0043,
      "mean_token_accuracy": 0.21825632601976394,
      "num_tokens": 98449741.0,
      "step": 47430
    },
    {
      "entropy": 5.443621253967285,
      "epoch": 4.302884615384615,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032745441369824485,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22616733312606813,
      "num_tokens": 98461068.0,
      "step": 47435
    },
    {
      "entropy": 5.482847404479981,
      "epoch": 4.303338171262699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032742294830423696,
      "loss": 4.981,
      "mean_token_accuracy": 0.21769258826971055,
      "num_tokens": 98472585.0,
      "step": 47440
    },
    {
      "entropy": 5.559524297714233,
      "epoch": 4.3037917271407835,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003273914818261092,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21959082931280136,
      "num_tokens": 98482874.0,
      "step": 47445
    },
    {
      "entropy": 5.511921787261963,
      "epoch": 4.304245283018868,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032736001426451216,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21891451627016068,
      "num_tokens": 98492995.0,
      "step": 47450
    },
    {
      "entropy": 5.516596937179566,
      "epoch": 4.304698838896952,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003273285456200966,
      "loss": 5.0178,
      "mean_token_accuracy": 0.21626180559396743,
      "num_tokens": 98503965.0,
      "step": 47455
    },
    {
      "entropy": 5.489794158935547,
      "epoch": 4.305152394775036,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003272970758935135,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21456556618213654,
      "num_tokens": 98515584.0,
      "step": 47460
    },
    {
      "entropy": 5.514266014099121,
      "epoch": 4.305605950653121,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032726560508541356,
      "loss": 5.019,
      "mean_token_accuracy": 0.21291071623563768,
      "num_tokens": 98527134.0,
      "step": 47465
    },
    {
      "entropy": 5.529429864883423,
      "epoch": 4.306059506531205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032723413319644756,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2125876411795616,
      "num_tokens": 98537189.0,
      "step": 47470
    },
    {
      "entropy": 5.494319057464599,
      "epoch": 4.306513062409289,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003272026602272663,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21573344022035598,
      "num_tokens": 98547803.0,
      "step": 47475
    },
    {
      "entropy": 5.4303606986999515,
      "epoch": 4.306966618287373,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032717118617852086,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2116665467619896,
      "num_tokens": 98558332.0,
      "step": 47480
    },
    {
      "entropy": 5.530064439773559,
      "epoch": 4.307420174165458,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003271397110508619,
      "loss": 5.0641,
      "mean_token_accuracy": 0.21110953241586686,
      "num_tokens": 98569046.0,
      "step": 47485
    },
    {
      "entropy": 5.508425378799439,
      "epoch": 4.307873730043541,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003271082348449405,
      "loss": 4.9592,
      "mean_token_accuracy": 0.2158171057701111,
      "num_tokens": 98579358.0,
      "step": 47490
    },
    {
      "entropy": 5.4978996276855465,
      "epoch": 4.308327285921625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003270767575614074,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21316667646169662,
      "num_tokens": 98589224.0,
      "step": 47495
    },
    {
      "entropy": 5.531364107131958,
      "epoch": 4.3087808417997095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003270452792009138,
      "loss": 5.0145,
      "mean_token_accuracy": 0.21842404305934907,
      "num_tokens": 98598954.0,
      "step": 47500
    },
    {
      "entropy": 5.432795763015747,
      "epoch": 4.309234397677794,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032701379976411043,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21587678045034409,
      "num_tokens": 98608813.0,
      "step": 47505
    },
    {
      "entropy": 5.459851551055908,
      "epoch": 4.309687953555878,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003269823192516486,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22135931104421616,
      "num_tokens": 98618746.0,
      "step": 47510
    },
    {
      "entropy": 5.463794946670532,
      "epoch": 4.310141509433962,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032695083766417913,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2212388500571251,
      "num_tokens": 98628804.0,
      "step": 47515
    },
    {
      "entropy": 5.441512107849121,
      "epoch": 4.310595065312047,
      "grad_norm": 1.0,
      "learning_rate": 0.000326919355002353,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22221851348876953,
      "num_tokens": 98640056.0,
      "step": 47520
    },
    {
      "entropy": 5.4546058177948,
      "epoch": 4.311048621190131,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032688787126682143,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2230391576886177,
      "num_tokens": 98651042.0,
      "step": 47525
    },
    {
      "entropy": 5.500126934051513,
      "epoch": 4.311502177068215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003268563864582354,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21734786927700042,
      "num_tokens": 98661405.0,
      "step": 47530
    },
    {
      "entropy": 5.527429676055908,
      "epoch": 4.3119557329462985,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032682490057724616,
      "loss": 5.01,
      "mean_token_accuracy": 0.21387053281068802,
      "num_tokens": 98672323.0,
      "step": 47535
    },
    {
      "entropy": 5.529012966156006,
      "epoch": 4.312409288824383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003267934136245047,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2116096779704094,
      "num_tokens": 98681431.0,
      "step": 47540
    },
    {
      "entropy": 5.370478773117066,
      "epoch": 4.312862844702467,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003267619256006623,
      "loss": 4.89,
      "mean_token_accuracy": 0.22985439300537108,
      "num_tokens": 98692345.0,
      "step": 47545
    },
    {
      "entropy": 5.470569181442261,
      "epoch": 4.313316400580551,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003267304365063701,
      "loss": 5.0045,
      "mean_token_accuracy": 0.21602013260126113,
      "num_tokens": 98703425.0,
      "step": 47550
    },
    {
      "entropy": 5.529816436767578,
      "epoch": 4.313769956458636,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003266989463422793,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21418153047561644,
      "num_tokens": 98714193.0,
      "step": 47555
    },
    {
      "entropy": 5.539413118362427,
      "epoch": 4.31422351233672,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032666745510904105,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22878233194351197,
      "num_tokens": 98723550.0,
      "step": 47560
    },
    {
      "entropy": 5.445914554595947,
      "epoch": 4.314677068214804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032663596280730673,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2144520491361618,
      "num_tokens": 98734075.0,
      "step": 47565
    },
    {
      "entropy": 5.453406286239624,
      "epoch": 4.315130624092888,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032660446943772755,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2185846447944641,
      "num_tokens": 98745447.0,
      "step": 47570
    },
    {
      "entropy": 5.528612232208252,
      "epoch": 4.315584179970973,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003265729750009547,
      "loss": 5.0269,
      "mean_token_accuracy": 0.21229286193847657,
      "num_tokens": 98756169.0,
      "step": 47575
    },
    {
      "entropy": 5.547678136825562,
      "epoch": 4.316037735849057,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003265414794976397,
      "loss": 5.0,
      "mean_token_accuracy": 0.2100517585873604,
      "num_tokens": 98767112.0,
      "step": 47580
    },
    {
      "entropy": 5.522531318664551,
      "epoch": 4.316491291727141,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003265099829284337,
      "loss": 5.0038,
      "mean_token_accuracy": 0.21730081737041473,
      "num_tokens": 98777533.0,
      "step": 47585
    },
    {
      "entropy": 5.532410192489624,
      "epoch": 4.316944847605225,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003264784852939881,
      "loss": 5.006,
      "mean_token_accuracy": 0.21106792986392975,
      "num_tokens": 98788101.0,
      "step": 47590
    },
    {
      "entropy": 5.599921321868896,
      "epoch": 4.317398403483309,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032644698659495437,
      "loss": 5.0033,
      "mean_token_accuracy": 0.21395463049411773,
      "num_tokens": 98799438.0,
      "step": 47595
    },
    {
      "entropy": 5.481949806213379,
      "epoch": 4.317851959361393,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032641548683198384,
      "loss": 4.9365,
      "mean_token_accuracy": 0.2161698743700981,
      "num_tokens": 98810325.0,
      "step": 47600
    },
    {
      "entropy": 5.43482813835144,
      "epoch": 4.318305515239477,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003263839860057279,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21784378737211227,
      "num_tokens": 98821897.0,
      "step": 47605
    },
    {
      "entropy": 5.489019393920898,
      "epoch": 4.318759071117562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000326352484116838,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2149947017431259,
      "num_tokens": 98833338.0,
      "step": 47610
    },
    {
      "entropy": 5.500288248062134,
      "epoch": 4.319212626995646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032632098116596576,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2141816422343254,
      "num_tokens": 98844308.0,
      "step": 47615
    },
    {
      "entropy": 5.4535962581634525,
      "epoch": 4.31966618287373,
      "grad_norm": 0.96875,
      "learning_rate": 0.00032628947715376247,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21895071268081664,
      "num_tokens": 98854491.0,
      "step": 47620
    },
    {
      "entropy": 5.495470237731934,
      "epoch": 4.320119738751814,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003262579720808797,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20798204094171524,
      "num_tokens": 98864702.0,
      "step": 47625
    },
    {
      "entropy": 5.486978435516358,
      "epoch": 4.320573294629899,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032622646594796905,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21948512494564057,
      "num_tokens": 98874682.0,
      "step": 47630
    },
    {
      "entropy": 5.533445549011231,
      "epoch": 4.321026850507983,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003261949587556821,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20557600557804107,
      "num_tokens": 98885130.0,
      "step": 47635
    },
    {
      "entropy": 5.49136791229248,
      "epoch": 4.321480406386067,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032616345050467016,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21868176609277726,
      "num_tokens": 98896398.0,
      "step": 47640
    },
    {
      "entropy": 5.451484823226929,
      "epoch": 4.321933962264151,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003261319411955852,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22239223569631578,
      "num_tokens": 98906571.0,
      "step": 47645
    },
    {
      "entropy": 5.458233499526978,
      "epoch": 4.322387518142235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003261004308290786,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22405171543359756,
      "num_tokens": 98916666.0,
      "step": 47650
    },
    {
      "entropy": 5.5144867420196535,
      "epoch": 4.322841074020319,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003260689194058021,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21598755866289138,
      "num_tokens": 98926599.0,
      "step": 47655
    },
    {
      "entropy": 5.4307066917419435,
      "epoch": 4.323294629898403,
      "grad_norm": 1.125,
      "learning_rate": 0.00032603740692640733,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2184246748685837,
      "num_tokens": 98936512.0,
      "step": 47660
    },
    {
      "entropy": 5.4870076179504395,
      "epoch": 4.323748185776488,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003260058933915459,
      "loss": 5.0251,
      "mean_token_accuracy": 0.20406153947114944,
      "num_tokens": 98946604.0,
      "step": 47665
    },
    {
      "entropy": 5.556351232528686,
      "epoch": 4.324201741654572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003259743788018697,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20663474053144454,
      "num_tokens": 98956257.0,
      "step": 47670
    },
    {
      "entropy": 5.510089683532715,
      "epoch": 4.324655297532656,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032594286315803026,
      "loss": 5.0032,
      "mean_token_accuracy": 0.21487257331609727,
      "num_tokens": 98966346.0,
      "step": 47675
    },
    {
      "entropy": 5.464644956588745,
      "epoch": 4.3251088534107405,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003259113464606795,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21484850943088532,
      "num_tokens": 98977410.0,
      "step": 47680
    },
    {
      "entropy": 5.5015500545501705,
      "epoch": 4.325562409288825,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003258798287104691,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2135014206171036,
      "num_tokens": 98987488.0,
      "step": 47685
    },
    {
      "entropy": 5.481344223022461,
      "epoch": 4.326015965166908,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032584830990805094,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22491728961467744,
      "num_tokens": 98996792.0,
      "step": 47690
    },
    {
      "entropy": 5.44023814201355,
      "epoch": 4.326469521044992,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032581679005407667,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22437380701303483,
      "num_tokens": 99007425.0,
      "step": 47695
    },
    {
      "entropy": 5.412955856323242,
      "epoch": 4.326923076923077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003257852691491983,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21733042001724243,
      "num_tokens": 99018275.0,
      "step": 47700
    },
    {
      "entropy": 5.48174204826355,
      "epoch": 4.327376632801161,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032575374719406755,
      "loss": 4.9882,
      "mean_token_accuracy": 0.2136937215924263,
      "num_tokens": 99028853.0,
      "step": 47705
    },
    {
      "entropy": 5.548014497756958,
      "epoch": 4.327830188679245,
      "grad_norm": 0.890625,
      "learning_rate": 0.00032572222418933644,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21046713143587112,
      "num_tokens": 99039423.0,
      "step": 47710
    },
    {
      "entropy": 5.506615400314331,
      "epoch": 4.3282837445573294,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003256907001356568,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20712383091449738,
      "num_tokens": 99050041.0,
      "step": 47715
    },
    {
      "entropy": 5.474574089050293,
      "epoch": 4.328737300435414,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003256591750336806,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2119929239153862,
      "num_tokens": 99062050.0,
      "step": 47720
    },
    {
      "entropy": 5.396959829330444,
      "epoch": 4.329190856313498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003256276488840596,
      "loss": 4.84,
      "mean_token_accuracy": 0.22202227711677552,
      "num_tokens": 99072337.0,
      "step": 47725
    },
    {
      "entropy": 5.409369230270386,
      "epoch": 4.329644412191582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032559612168744613,
      "loss": 4.8642,
      "mean_token_accuracy": 0.21752171516418456,
      "num_tokens": 99082468.0,
      "step": 47730
    },
    {
      "entropy": 5.4935657501220705,
      "epoch": 4.3300979680696665,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003255645934444918,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21075666844844818,
      "num_tokens": 99091757.0,
      "step": 47735
    },
    {
      "entropy": 5.522720861434936,
      "epoch": 4.330551523947751,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003255330641558489,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20915383249521255,
      "num_tokens": 99103192.0,
      "step": 47740
    },
    {
      "entropy": 5.392834424972534,
      "epoch": 4.331005079825834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003255015338221693,
      "loss": 4.8514,
      "mean_token_accuracy": 0.21996861696243286,
      "num_tokens": 99112660.0,
      "step": 47745
    },
    {
      "entropy": 5.496247577667236,
      "epoch": 4.331458635703918,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00032547000244410515,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21376363337039947,
      "num_tokens": 99123867.0,
      "step": 47750
    },
    {
      "entropy": 5.509067010879517,
      "epoch": 4.331912191582003,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032543847002230845,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21151411533355713,
      "num_tokens": 99134445.0,
      "step": 47755
    },
    {
      "entropy": 5.415577554702759,
      "epoch": 4.332365747460087,
      "grad_norm": 1.125,
      "learning_rate": 0.0003254069365574313,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22331182658672333,
      "num_tokens": 99143483.0,
      "step": 47760
    },
    {
      "entropy": 5.476224040985107,
      "epoch": 4.332819303338171,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003253754020501259,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22199078202247619,
      "num_tokens": 99154514.0,
      "step": 47765
    },
    {
      "entropy": 5.513947010040283,
      "epoch": 4.3332728592162555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032534386650104444,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20909756124019624,
      "num_tokens": 99165969.0,
      "step": 47770
    },
    {
      "entropy": 5.394663333892822,
      "epoch": 4.33372641509434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003253123299108388,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2245722532272339,
      "num_tokens": 99176595.0,
      "step": 47775
    },
    {
      "entropy": 5.489801836013794,
      "epoch": 4.334179970972424,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003252807922801615,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21284951865673066,
      "num_tokens": 99186916.0,
      "step": 47780
    },
    {
      "entropy": 5.531912183761596,
      "epoch": 4.334633526850508,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003252492536096646,
      "loss": 5.0997,
      "mean_token_accuracy": 0.2066388174891472,
      "num_tokens": 99197892.0,
      "step": 47785
    },
    {
      "entropy": 5.560431098937988,
      "epoch": 4.3350870827285926,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003252177139000003,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21826505661010742,
      "num_tokens": 99208373.0,
      "step": 47790
    },
    {
      "entropy": 5.524578714370728,
      "epoch": 4.335540638606676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003251861731518209,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21934956610202788,
      "num_tokens": 99218788.0,
      "step": 47795
    },
    {
      "entropy": 5.558471918106079,
      "epoch": 4.33599419448476,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032515463136577864,
      "loss": 5.0242,
      "mean_token_accuracy": 0.2115597367286682,
      "num_tokens": 99228653.0,
      "step": 47800
    },
    {
      "entropy": 5.463826274871826,
      "epoch": 4.3364477503628445,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032512308854252577,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22127263844013215,
      "num_tokens": 99239925.0,
      "step": 47805
    },
    {
      "entropy": 5.404535865783691,
      "epoch": 4.336901306240929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003250915446827147,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2269389659166336,
      "num_tokens": 99250091.0,
      "step": 47810
    },
    {
      "entropy": 5.457209348678589,
      "epoch": 4.337354862119013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003250599997869977,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20189599841833114,
      "num_tokens": 99260799.0,
      "step": 47815
    },
    {
      "entropy": 5.435519456863403,
      "epoch": 4.337808417997097,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003250284538560272,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2250112771987915,
      "num_tokens": 99270338.0,
      "step": 47820
    },
    {
      "entropy": 5.471937751770019,
      "epoch": 4.3382619738751815,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003249969068904554,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21390779167413712,
      "num_tokens": 99280419.0,
      "step": 47825
    },
    {
      "entropy": 5.547903442382813,
      "epoch": 4.338715529753266,
      "grad_norm": 0.953125,
      "learning_rate": 0.00032496535889093496,
      "loss": 5.0199,
      "mean_token_accuracy": 0.21073633432388306,
      "num_tokens": 99291590.0,
      "step": 47830
    },
    {
      "entropy": 5.515287828445435,
      "epoch": 4.33916908563135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032493380985811814,
      "loss": 5.0003,
      "mean_token_accuracy": 0.2123573675751686,
      "num_tokens": 99301987.0,
      "step": 47835
    },
    {
      "entropy": 5.554364824295044,
      "epoch": 4.339622641509434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032490225979265734,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21959315240383148,
      "num_tokens": 99312231.0,
      "step": 47840
    },
    {
      "entropy": 5.5085773944854735,
      "epoch": 4.340076197387518,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032487070869520517,
      "loss": 4.8992,
      "mean_token_accuracy": 0.21850819140672684,
      "num_tokens": 99322784.0,
      "step": 47845
    },
    {
      "entropy": 5.508476448059082,
      "epoch": 4.340529753265602,
      "grad_norm": 1.03125,
      "learning_rate": 0.000324839156566414,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21251858472824098,
      "num_tokens": 99333989.0,
      "step": 47850
    },
    {
      "entropy": 5.440493106842041,
      "epoch": 4.340983309143686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003248076034069364,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2186536967754364,
      "num_tokens": 99345034.0,
      "step": 47855
    },
    {
      "entropy": 5.4360212802886965,
      "epoch": 4.3414368650217705,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003247760492174248,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21840727031230928,
      "num_tokens": 99356062.0,
      "step": 47860
    },
    {
      "entropy": 5.478455638885498,
      "epoch": 4.341890420899855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003247444939985319,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22144026905298234,
      "num_tokens": 99365315.0,
      "step": 47865
    },
    {
      "entropy": 5.529478883743286,
      "epoch": 4.342343976777939,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032471293775091013,
      "loss": 4.942,
      "mean_token_accuracy": 0.22632029801607131,
      "num_tokens": 99374897.0,
      "step": 47870
    },
    {
      "entropy": 5.436788749694824,
      "epoch": 4.342797532656023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032468138047521216,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2210872381925583,
      "num_tokens": 99383624.0,
      "step": 47875
    },
    {
      "entropy": 5.461963891983032,
      "epoch": 4.343251088534108,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003246498221720906,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21287599503993987,
      "num_tokens": 99393791.0,
      "step": 47880
    },
    {
      "entropy": 5.502488327026367,
      "epoch": 4.343704644412192,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000324618262842198,
      "loss": 4.9776,
      "mean_token_accuracy": 0.21972359865903854,
      "num_tokens": 99403687.0,
      "step": 47885
    },
    {
      "entropy": 5.535463905334472,
      "epoch": 4.344158200290276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032458670248618714,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2122011661529541,
      "num_tokens": 99413411.0,
      "step": 47890
    },
    {
      "entropy": 5.530371475219726,
      "epoch": 4.34461175616836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003245551411047106,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2112942636013031,
      "num_tokens": 99425563.0,
      "step": 47895
    },
    {
      "entropy": 5.460479736328125,
      "epoch": 4.345065312046444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032452357869842117,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22916054129600524,
      "num_tokens": 99436297.0,
      "step": 47900
    },
    {
      "entropy": 5.492803907394409,
      "epoch": 4.345518867924528,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003244920152679715,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21510404944419861,
      "num_tokens": 99445745.0,
      "step": 47905
    },
    {
      "entropy": 5.44848256111145,
      "epoch": 4.345972423802612,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032446045081401426,
      "loss": 4.9715,
      "mean_token_accuracy": 0.21287051290273667,
      "num_tokens": 99456915.0,
      "step": 47910
    },
    {
      "entropy": 5.522615051269531,
      "epoch": 4.3464259796806965,
      "grad_norm": 1.0,
      "learning_rate": 0.0003244288853372024,
      "loss": 4.9702,
      "mean_token_accuracy": 0.2147314041852951,
      "num_tokens": 99466778.0,
      "step": 47915
    },
    {
      "entropy": 5.5060577392578125,
      "epoch": 4.346879535558781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003243973188381885,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2145090237259865,
      "num_tokens": 99475984.0,
      "step": 47920
    },
    {
      "entropy": 5.398883247375489,
      "epoch": 4.347333091436865,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032436575131762553,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21393485069274903,
      "num_tokens": 99486238.0,
      "step": 47925
    },
    {
      "entropy": 5.600556373596191,
      "epoch": 4.347786647314949,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032433418277616623,
      "loss": 5.0413,
      "mean_token_accuracy": 0.21073847115039826,
      "num_tokens": 99497806.0,
      "step": 47930
    },
    {
      "entropy": 5.47787561416626,
      "epoch": 4.348240203193034,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003243026132144635,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21324743777513505,
      "num_tokens": 99507214.0,
      "step": 47935
    },
    {
      "entropy": 5.4879075050354,
      "epoch": 4.348693759071118,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003242710426331701,
      "loss": 4.9798,
      "mean_token_accuracy": 0.215549236536026,
      "num_tokens": 99517025.0,
      "step": 47940
    },
    {
      "entropy": 5.515346193313599,
      "epoch": 4.349147314949202,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032423947103293893,
      "loss": 4.9912,
      "mean_token_accuracy": 0.21699889600276948,
      "num_tokens": 99526972.0,
      "step": 47945
    },
    {
      "entropy": 5.538156127929687,
      "epoch": 4.3496008708272855,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032420789841442306,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2101664960384369,
      "num_tokens": 99538583.0,
      "step": 47950
    },
    {
      "entropy": 5.426387739181519,
      "epoch": 4.35005442670537,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032417632477827526,
      "loss": 4.8772,
      "mean_token_accuracy": 0.2187370166182518,
      "num_tokens": 99549178.0,
      "step": 47955
    },
    {
      "entropy": 5.430292463302612,
      "epoch": 4.350507982583454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032414475012514847,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22087093442678452,
      "num_tokens": 99558491.0,
      "step": 47960
    },
    {
      "entropy": 5.490369033813477,
      "epoch": 4.350961538461538,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003241131744556958,
      "loss": 4.984,
      "mean_token_accuracy": 0.21198045909404756,
      "num_tokens": 99568711.0,
      "step": 47965
    },
    {
      "entropy": 5.448770475387573,
      "epoch": 4.351415094339623,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003240815977705702,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21456683725118636,
      "num_tokens": 99578329.0,
      "step": 47970
    },
    {
      "entropy": 5.413377857208252,
      "epoch": 4.351868650217707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032405002007042453,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22287399172782899,
      "num_tokens": 99588646.0,
      "step": 47975
    },
    {
      "entropy": 5.504530429840088,
      "epoch": 4.352322206095791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032401844135591213,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2098693311214447,
      "num_tokens": 99598780.0,
      "step": 47980
    },
    {
      "entropy": 5.498278617858887,
      "epoch": 4.352775761973875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003239868616276857,
      "loss": 4.9636,
      "mean_token_accuracy": 0.22192862182855605,
      "num_tokens": 99609528.0,
      "step": 47985
    },
    {
      "entropy": 5.400754308700561,
      "epoch": 4.35322931785196,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003239552808863986,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22036208361387252,
      "num_tokens": 99619928.0,
      "step": 47990
    },
    {
      "entropy": 5.389384984970093,
      "epoch": 4.353682873730044,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032392369913270376,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22197439819574355,
      "num_tokens": 99630135.0,
      "step": 47995
    },
    {
      "entropy": 5.508530139923096,
      "epoch": 4.354136429608127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003238921163672544,
      "loss": 5.0579,
      "mean_token_accuracy": 0.19914040863513946,
      "num_tokens": 99641557.0,
      "step": 48000
    },
    {
      "epoch": 4.354136429608127,
      "eval_entropy": 5.2485662910193875,
      "eval_loss": 5.046014308929443,
      "eval_mean_token_accuracy": 0.21949108014582777,
      "eval_num_tokens": 99641557.0,
      "eval_runtime": 26.6317,
      "eval_samples_per_second": 1327.743,
      "eval_steps_per_second": 165.968,
      "step": 48000
    },
    {
      "entropy": 5.524392366409302,
      "epoch": 4.354589985486212,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003238605325907035,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21730905622243882,
      "num_tokens": 99651244.0,
      "step": 48005
    },
    {
      "entropy": 5.511731338500977,
      "epoch": 4.355043541364296,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032382894780370446,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21401729434728622,
      "num_tokens": 99662317.0,
      "step": 48010
    },
    {
      "entropy": 5.433379316329956,
      "epoch": 4.35549709724238,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032379736200691036,
      "loss": 4.889,
      "mean_token_accuracy": 0.22313737273216247,
      "num_tokens": 99671624.0,
      "step": 48015
    },
    {
      "entropy": 5.479178810119629,
      "epoch": 4.355950653120464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032376577520097427,
      "loss": 5.0415,
      "mean_token_accuracy": 0.21049033254384994,
      "num_tokens": 99682076.0,
      "step": 48020
    },
    {
      "entropy": 5.477306175231933,
      "epoch": 4.356404208998549,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032373418738654953,
      "loss": 4.93,
      "mean_token_accuracy": 0.22125840932130814,
      "num_tokens": 99692076.0,
      "step": 48025
    },
    {
      "entropy": 5.488963031768799,
      "epoch": 4.356857764876633,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003237025985642895,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22104749977588653,
      "num_tokens": 99701455.0,
      "step": 48030
    },
    {
      "entropy": 5.502605390548706,
      "epoch": 4.357311320754717,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032367100873484725,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2154761239886284,
      "num_tokens": 99712034.0,
      "step": 48035
    },
    {
      "entropy": 5.462673187255859,
      "epoch": 4.357764876632801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000323639417898876,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21328754276037215,
      "num_tokens": 99722965.0,
      "step": 48040
    },
    {
      "entropy": 5.376438331604004,
      "epoch": 4.358218432510886,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003236078260570293,
      "loss": 4.8553,
      "mean_token_accuracy": 0.21794369965791702,
      "num_tokens": 99732847.0,
      "step": 48045
    },
    {
      "entropy": 5.413998079299927,
      "epoch": 4.35867198838897,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003235762332099604,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22889200150966643,
      "num_tokens": 99743639.0,
      "step": 48050
    },
    {
      "entropy": 5.431378126144409,
      "epoch": 4.359125544267053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003235446393583227,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23276870846748351,
      "num_tokens": 99753374.0,
      "step": 48055
    },
    {
      "entropy": 5.45464563369751,
      "epoch": 4.359579100145138,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003235130445027693,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22032787799835205,
      "num_tokens": 99763361.0,
      "step": 48060
    },
    {
      "entropy": 5.523088836669922,
      "epoch": 4.360032656023222,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032348144864395383,
      "loss": 5.0269,
      "mean_token_accuracy": 0.21302133947610855,
      "num_tokens": 99773813.0,
      "step": 48065
    },
    {
      "entropy": 5.502657794952393,
      "epoch": 4.360486211901306,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003234498517825297,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21664670407772063,
      "num_tokens": 99784812.0,
      "step": 48070
    },
    {
      "entropy": 5.475845766067505,
      "epoch": 4.36093976777939,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00032341825391915026,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21883524656295777,
      "num_tokens": 99795197.0,
      "step": 48075
    },
    {
      "entropy": 5.417968273162842,
      "epoch": 4.361393323657475,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000323386655054469,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21940976828336717,
      "num_tokens": 99805883.0,
      "step": 48080
    },
    {
      "entropy": 5.507496309280396,
      "epoch": 4.361846879535559,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003233550551891394,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21755834221839904,
      "num_tokens": 99816940.0,
      "step": 48085
    },
    {
      "entropy": 5.450926065444946,
      "epoch": 4.362300435413643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032332345432381493,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2151022419333458,
      "num_tokens": 99827926.0,
      "step": 48090
    },
    {
      "entropy": 5.397157859802246,
      "epoch": 4.3627539912917275,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003232918524591491,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22400778234004975,
      "num_tokens": 99837900.0,
      "step": 48095
    },
    {
      "entropy": 5.509851694107056,
      "epoch": 4.363207547169811,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003232602495957955,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2080420583486557,
      "num_tokens": 99848655.0,
      "step": 48100
    },
    {
      "entropy": 5.429569911956787,
      "epoch": 4.363661103047895,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00032322864573440757,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22293040454387664,
      "num_tokens": 99859715.0,
      "step": 48105
    },
    {
      "entropy": 5.461143541336059,
      "epoch": 4.364114658925979,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003231970408756389,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2225741535425186,
      "num_tokens": 99870175.0,
      "step": 48110
    },
    {
      "entropy": 5.537932443618774,
      "epoch": 4.364568214804064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003231654350201432,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21086916625499724,
      "num_tokens": 99879794.0,
      "step": 48115
    },
    {
      "entropy": 5.523962116241455,
      "epoch": 4.365021770682148,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003231338281685741,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22787429690361022,
      "num_tokens": 99889418.0,
      "step": 48120
    },
    {
      "entropy": 5.526042842864991,
      "epoch": 4.365475326560232,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003231022203215851,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21029845923185347,
      "num_tokens": 99898741.0,
      "step": 48125
    },
    {
      "entropy": 5.422848033905029,
      "epoch": 4.3659288824383164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032307061147982994,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21648626029491425,
      "num_tokens": 99909167.0,
      "step": 48130
    },
    {
      "entropy": 5.424997043609619,
      "epoch": 4.366382438316401,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003230390016439623,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21921151280403137,
      "num_tokens": 99918786.0,
      "step": 48135
    },
    {
      "entropy": 5.438978052139282,
      "epoch": 4.366835994194485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003230073908146359,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21366580724716186,
      "num_tokens": 99928758.0,
      "step": 48140
    },
    {
      "entropy": 5.469822835922241,
      "epoch": 4.367289550072569,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003229757789925044,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21724718064069748,
      "num_tokens": 99939721.0,
      "step": 48145
    },
    {
      "entropy": 5.470340347290039,
      "epoch": 4.3677431059506535,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003229441661782215,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22202271074056626,
      "num_tokens": 99949308.0,
      "step": 48150
    },
    {
      "entropy": 5.472623729705811,
      "epoch": 4.368196661828737,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032291255237244104,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21404284834861756,
      "num_tokens": 99960284.0,
      "step": 48155
    },
    {
      "entropy": 5.397556591033935,
      "epoch": 4.368650217706821,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032288093757581684,
      "loss": 4.8257,
      "mean_token_accuracy": 0.22234335988759996,
      "num_tokens": 99970701.0,
      "step": 48160
    },
    {
      "entropy": 5.457958316802978,
      "epoch": 4.369103773584905,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003228493217890026,
      "loss": 4.8959,
      "mean_token_accuracy": 0.21668666899204253,
      "num_tokens": 99981362.0,
      "step": 48165
    },
    {
      "entropy": 5.4496180534362795,
      "epoch": 4.36955732946299,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003228177050126523,
      "loss": 4.951,
      "mean_token_accuracy": 0.21857725381851195,
      "num_tokens": 99991383.0,
      "step": 48170
    },
    {
      "entropy": 5.456042957305908,
      "epoch": 4.370010885341074,
      "grad_norm": 0.9375,
      "learning_rate": 0.00032278608724741956,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21620429307222366,
      "num_tokens": 100001176.0,
      "step": 48175
    },
    {
      "entropy": 5.473151588439942,
      "epoch": 4.370464441219158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003227544684939584,
      "loss": 4.889,
      "mean_token_accuracy": 0.22793350666761397,
      "num_tokens": 100011467.0,
      "step": 48180
    },
    {
      "entropy": 5.477672481536866,
      "epoch": 4.3709179970972425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003227228487529226,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22078366428613663,
      "num_tokens": 100020976.0,
      "step": 48185
    },
    {
      "entropy": 5.417583656311035,
      "epoch": 4.371371552975327,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003226912280249662,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21580397039651872,
      "num_tokens": 100032250.0,
      "step": 48190
    },
    {
      "entropy": 5.475901699066162,
      "epoch": 4.371825108853411,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000322659606310743,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21794375777244568,
      "num_tokens": 100043383.0,
      "step": 48195
    },
    {
      "entropy": 5.435182189941406,
      "epoch": 4.372278664731495,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032262798361090705,
      "loss": 4.8597,
      "mean_token_accuracy": 0.22940587103366852,
      "num_tokens": 100053396.0,
      "step": 48200
    },
    {
      "entropy": 5.442156505584717,
      "epoch": 4.37273222060958,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003225963599261123,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21830297708511354,
      "num_tokens": 100064159.0,
      "step": 48205
    },
    {
      "entropy": 5.430845022201538,
      "epoch": 4.373185776487663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032256473525701257,
      "loss": 4.9104,
      "mean_token_accuracy": 0.22684472054243088,
      "num_tokens": 100074722.0,
      "step": 48210
    },
    {
      "entropy": 5.41517014503479,
      "epoch": 4.373639332365747,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032253310960426215,
      "loss": 4.9035,
      "mean_token_accuracy": 0.23023841083049773,
      "num_tokens": 100085067.0,
      "step": 48215
    },
    {
      "entropy": 5.477180671691895,
      "epoch": 4.3740928882438315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032250148296851484,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22417379021644593,
      "num_tokens": 100095806.0,
      "step": 48220
    },
    {
      "entropy": 5.5716877460479735,
      "epoch": 4.374546444121916,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003224698553504247,
      "loss": 5.0103,
      "mean_token_accuracy": 0.212394680082798,
      "num_tokens": 100106069.0,
      "step": 48225
    },
    {
      "entropy": 5.582152509689331,
      "epoch": 4.375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032243822675064585,
      "loss": 5.0047,
      "mean_token_accuracy": 0.21245371997356416,
      "num_tokens": 100117136.0,
      "step": 48230
    },
    {
      "entropy": 5.491768836975098,
      "epoch": 4.375453555878084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003224065971698324,
      "loss": 4.9486,
      "mean_token_accuracy": 0.20403512567281723,
      "num_tokens": 100126757.0,
      "step": 48235
    },
    {
      "entropy": 5.579689025878906,
      "epoch": 4.3759071117561685,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003223749666086385,
      "loss": 5.0841,
      "mean_token_accuracy": 0.21036505252122878,
      "num_tokens": 100138431.0,
      "step": 48240
    },
    {
      "entropy": 5.474396705627441,
      "epoch": 4.376360667634253,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003223433350677182,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2157072678208351,
      "num_tokens": 100148919.0,
      "step": 48245
    },
    {
      "entropy": 5.571362018585205,
      "epoch": 4.376814223512337,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003223117025477256,
      "loss": 5.0869,
      "mean_token_accuracy": 0.20911828130483628,
      "num_tokens": 100159175.0,
      "step": 48250
    },
    {
      "entropy": 5.476064395904541,
      "epoch": 4.37726777939042,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003222800690493151,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21602344512939453,
      "num_tokens": 100169451.0,
      "step": 48255
    },
    {
      "entropy": 5.545414400100708,
      "epoch": 4.377721335268505,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003222484345731405,
      "loss": 4.9476,
      "mean_token_accuracy": 0.21900524497032164,
      "num_tokens": 100180077.0,
      "step": 48260
    },
    {
      "entropy": 5.4482118606567385,
      "epoch": 4.378174891146589,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032221679911985634,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21896248459815978,
      "num_tokens": 100190175.0,
      "step": 48265
    },
    {
      "entropy": 5.3634593963623045,
      "epoch": 4.378628447024673,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032218516269011676,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22490948140621186,
      "num_tokens": 100200285.0,
      "step": 48270
    },
    {
      "entropy": 5.354559803009034,
      "epoch": 4.3790820029027575,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032215352528457597,
      "loss": 4.783,
      "mean_token_accuracy": 0.22658265829086305,
      "num_tokens": 100210685.0,
      "step": 48275
    },
    {
      "entropy": 5.460714292526245,
      "epoch": 4.379535558780842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032212188690388825,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2284626245498657,
      "num_tokens": 100220486.0,
      "step": 48280
    },
    {
      "entropy": 5.423684406280517,
      "epoch": 4.379989114658926,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032209024754870796,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22403409481048583,
      "num_tokens": 100229007.0,
      "step": 48285
    },
    {
      "entropy": 5.435066652297974,
      "epoch": 4.38044267053701,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003220586072196893,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22044081389904022,
      "num_tokens": 100240363.0,
      "step": 48290
    },
    {
      "entropy": 5.412042331695557,
      "epoch": 4.380896226415095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032202696591748667,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22359195947647095,
      "num_tokens": 100251193.0,
      "step": 48295
    },
    {
      "entropy": 5.46091594696045,
      "epoch": 4.381349782293179,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032199532364275436,
      "loss": 4.9139,
      "mean_token_accuracy": 0.22328902930021285,
      "num_tokens": 100261248.0,
      "step": 48300
    },
    {
      "entropy": 5.464053726196289,
      "epoch": 4.381803338171263,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003219636803961468,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2128388375043869,
      "num_tokens": 100271174.0,
      "step": 48305
    },
    {
      "entropy": 5.444920873641967,
      "epoch": 4.3822568940493465,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032193203617831835,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21414474099874498,
      "num_tokens": 100282428.0,
      "step": 48310
    },
    {
      "entropy": 5.503177165985107,
      "epoch": 4.382710449927431,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032190039098992345,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21604271829128266,
      "num_tokens": 100292559.0,
      "step": 48315
    },
    {
      "entropy": 5.437931299209595,
      "epoch": 4.383164005805515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003218687448316166,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22691166549921035,
      "num_tokens": 100303956.0,
      "step": 48320
    },
    {
      "entropy": 5.398964357376099,
      "epoch": 4.383617561683599,
      "grad_norm": 0.90625,
      "learning_rate": 0.00032183709770405204,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21735940128564835,
      "num_tokens": 100315281.0,
      "step": 48325
    },
    {
      "entropy": 5.494449186325073,
      "epoch": 4.3840711175616836,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003218054496078844,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21374516040086747,
      "num_tokens": 100326761.0,
      "step": 48330
    },
    {
      "entropy": 5.489939403533936,
      "epoch": 4.384524673439768,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032177380054376813,
      "loss": 4.9821,
      "mean_token_accuracy": 0.21834274679422377,
      "num_tokens": 100337219.0,
      "step": 48335
    },
    {
      "entropy": 5.469893980026245,
      "epoch": 4.384978229317852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032174215051235776,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2223561927676201,
      "num_tokens": 100347135.0,
      "step": 48340
    },
    {
      "entropy": 5.358528327941895,
      "epoch": 4.385431785195936,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032171049951430775,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2342589884996414,
      "num_tokens": 100357476.0,
      "step": 48345
    },
    {
      "entropy": 5.449358558654785,
      "epoch": 4.385885341074021,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032167884755027285,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21451250314712525,
      "num_tokens": 100367860.0,
      "step": 48350
    },
    {
      "entropy": 5.484124088287354,
      "epoch": 4.386338896952105,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003216471946209074,
      "loss": 4.9007,
      "mean_token_accuracy": 0.22044722139835357,
      "num_tokens": 100378613.0,
      "step": 48355
    },
    {
      "entropy": 5.444883155822754,
      "epoch": 4.386792452830189,
      "grad_norm": 1.09375,
      "learning_rate": 0.000321615540726866,
      "loss": 4.9664,
      "mean_token_accuracy": 0.218281652033329,
      "num_tokens": 100389499.0,
      "step": 48360
    },
    {
      "entropy": 5.495291996002197,
      "epoch": 4.3872460087082725,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00032158388586880346,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21796678751707077,
      "num_tokens": 100400538.0,
      "step": 48365
    },
    {
      "entropy": 5.479336738586426,
      "epoch": 4.387699564586357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032155223004737423,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2226123809814453,
      "num_tokens": 100411528.0,
      "step": 48370
    },
    {
      "entropy": 5.528587007522583,
      "epoch": 4.388153120464441,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032152057326323305,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21151998341083528,
      "num_tokens": 100421349.0,
      "step": 48375
    },
    {
      "entropy": 5.419046449661255,
      "epoch": 4.388606676342525,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032148891551703445,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22053847461938858,
      "num_tokens": 100431132.0,
      "step": 48380
    },
    {
      "entropy": 5.4189342021942135,
      "epoch": 4.38906023222061,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003214572568094334,
      "loss": 4.9427,
      "mean_token_accuracy": 0.22595446258783342,
      "num_tokens": 100441436.0,
      "step": 48385
    },
    {
      "entropy": 5.4131989002227785,
      "epoch": 4.389513788098694,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003214255971410843,
      "loss": 4.7913,
      "mean_token_accuracy": 0.22772494852542877,
      "num_tokens": 100451711.0,
      "step": 48390
    },
    {
      "entropy": 5.57384762763977,
      "epoch": 4.389967343976778,
      "grad_norm": 1.015625,
      "learning_rate": 0.000321393936512642,
      "loss": 5.0724,
      "mean_token_accuracy": 0.21372431814670562,
      "num_tokens": 100461177.0,
      "step": 48395
    },
    {
      "entropy": 5.430560302734375,
      "epoch": 4.390420899854862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032136227492476145,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2166861966252327,
      "num_tokens": 100471391.0,
      "step": 48400
    },
    {
      "entropy": 5.492292308807373,
      "epoch": 4.390874455732947,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003213306123780971,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22255857437849044,
      "num_tokens": 100480747.0,
      "step": 48405
    },
    {
      "entropy": 5.513067054748535,
      "epoch": 4.39132801161103,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032129894887330396,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2200464129447937,
      "num_tokens": 100491451.0,
      "step": 48410
    },
    {
      "entropy": 5.5342343807220455,
      "epoch": 4.391781567489114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003212672844110367,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21403357684612273,
      "num_tokens": 100501801.0,
      "step": 48415
    },
    {
      "entropy": 5.394865655899048,
      "epoch": 4.392235123367199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032123561899195026,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2243402808904648,
      "num_tokens": 100511857.0,
      "step": 48420
    },
    {
      "entropy": 5.520209455490113,
      "epoch": 4.392688679245283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003212039526166994,
      "loss": 4.949,
      "mean_token_accuracy": 0.21649111360311507,
      "num_tokens": 100523654.0,
      "step": 48425
    },
    {
      "entropy": 5.5512909412384035,
      "epoch": 4.393142235123367,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032117228528593907,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2169978991150856,
      "num_tokens": 100533030.0,
      "step": 48430
    },
    {
      "entropy": 5.518808650970459,
      "epoch": 4.393595791001451,
      "grad_norm": 1.0625,
      "learning_rate": 0.000321140617000324,
      "loss": 4.956,
      "mean_token_accuracy": 0.213572858273983,
      "num_tokens": 100543414.0,
      "step": 48435
    },
    {
      "entropy": 5.5146719455719,
      "epoch": 4.394049346879536,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032110894776050936,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21956171840429306,
      "num_tokens": 100553303.0,
      "step": 48440
    },
    {
      "entropy": 5.4315526485443115,
      "epoch": 4.39450290275762,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032107727756714984,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21867032796144487,
      "num_tokens": 100563030.0,
      "step": 48445
    },
    {
      "entropy": 5.523464918136597,
      "epoch": 4.394956458635704,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003210456064209005,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21012945622205734,
      "num_tokens": 100573216.0,
      "step": 48450
    },
    {
      "entropy": 5.430750799179077,
      "epoch": 4.395410014513788,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00032101393432241635,
      "loss": 4.8463,
      "mean_token_accuracy": 0.22071876078844072,
      "num_tokens": 100582646.0,
      "step": 48455
    },
    {
      "entropy": 5.453864574432373,
      "epoch": 4.395863570391873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032098226127235215,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20933522135019303,
      "num_tokens": 100593689.0,
      "step": 48460
    },
    {
      "entropy": 5.489145708084107,
      "epoch": 4.396317126269956,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003209505872713632,
      "loss": 5.0055,
      "mean_token_accuracy": 0.212027470767498,
      "num_tokens": 100604721.0,
      "step": 48465
    },
    {
      "entropy": 5.509261608123779,
      "epoch": 4.39677068214804,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00032091891232010434,
      "loss": 4.9939,
      "mean_token_accuracy": 0.21880888044834138,
      "num_tokens": 100614684.0,
      "step": 48470
    },
    {
      "entropy": 5.474276447296143,
      "epoch": 4.397224238026125,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032088723641923066,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21926482617855073,
      "num_tokens": 100624434.0,
      "step": 48475
    },
    {
      "entropy": 5.548297309875489,
      "epoch": 4.397677793904209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032085555956939726,
      "loss": 5.0787,
      "mean_token_accuracy": 0.2021039754152298,
      "num_tokens": 100635656.0,
      "step": 48480
    },
    {
      "entropy": 5.376196336746216,
      "epoch": 4.398131349782293,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003208238817712593,
      "loss": 4.8065,
      "mean_token_accuracy": 0.22818323075771332,
      "num_tokens": 100645749.0,
      "step": 48485
    },
    {
      "entropy": 5.425412893295288,
      "epoch": 4.398584905660377,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003207922030254717,
      "loss": 4.901,
      "mean_token_accuracy": 0.22236181944608688,
      "num_tokens": 100655814.0,
      "step": 48490
    },
    {
      "entropy": 5.4179627895355225,
      "epoch": 4.399038461538462,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032076052333268964,
      "loss": 4.893,
      "mean_token_accuracy": 0.21745591461658478,
      "num_tokens": 100665579.0,
      "step": 48495
    },
    {
      "entropy": 5.580921173095703,
      "epoch": 4.399492017416546,
      "grad_norm": 1.125,
      "learning_rate": 0.0003207288426935683,
      "loss": 5.0982,
      "mean_token_accuracy": 0.2083563596010208,
      "num_tokens": 100676421.0,
      "step": 48500
    },
    {
      "entropy": 5.483607769012451,
      "epoch": 4.39994557329463,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003206971611087629,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22385682612657548,
      "num_tokens": 100686689.0,
      "step": 48505
    },
    {
      "entropy": 5.5730156898498535,
      "epoch": 4.4003991291727145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003206654785789285,
      "loss": 5.0305,
      "mean_token_accuracy": 0.21707392036914824,
      "num_tokens": 100697324.0,
      "step": 48510
    },
    {
      "entropy": 5.38490514755249,
      "epoch": 4.400852685050798,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032063379510472036,
      "loss": 4.8434,
      "mean_token_accuracy": 0.2286358579993248,
      "num_tokens": 100706003.0,
      "step": 48515
    },
    {
      "entropy": 5.516802883148193,
      "epoch": 4.401306240928882,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003206021106867938,
      "loss": 4.985,
      "mean_token_accuracy": 0.22207142114639283,
      "num_tokens": 100716755.0,
      "step": 48520
    },
    {
      "entropy": 5.4788806438446045,
      "epoch": 4.401759796806966,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003205704253258039,
      "loss": 4.9927,
      "mean_token_accuracy": 0.21299874484539033,
      "num_tokens": 100727846.0,
      "step": 48525
    },
    {
      "entropy": 5.456745672225952,
      "epoch": 4.402213352685051,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000320538739022406,
      "loss": 4.8929,
      "mean_token_accuracy": 0.21357960850000382,
      "num_tokens": 100738293.0,
      "step": 48530
    },
    {
      "entropy": 5.460463190078736,
      "epoch": 4.402666908563135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003205070517772554,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2201475292444229,
      "num_tokens": 100748031.0,
      "step": 48535
    },
    {
      "entropy": 5.414367723464966,
      "epoch": 4.403120464441219,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003204753635910073,
      "loss": 4.912,
      "mean_token_accuracy": 0.22462691366672516,
      "num_tokens": 100758825.0,
      "step": 48540
    },
    {
      "entropy": 5.432081174850464,
      "epoch": 4.4035740203193035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032044367446431714,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22498926222324372,
      "num_tokens": 100768226.0,
      "step": 48545
    },
    {
      "entropy": 5.435732173919678,
      "epoch": 4.404027576197388,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032041198439784025,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22308627516031265,
      "num_tokens": 100778125.0,
      "step": 48550
    },
    {
      "entropy": 5.496981954574585,
      "epoch": 4.404481132075472,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003203802933922319,
      "loss": 4.9259,
      "mean_token_accuracy": 0.22322166115045547,
      "num_tokens": 100788670.0,
      "step": 48555
    },
    {
      "entropy": 5.512003993988037,
      "epoch": 4.404934687953556,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032034860144814756,
      "loss": 4.9706,
      "mean_token_accuracy": 0.2086901694536209,
      "num_tokens": 100800077.0,
      "step": 48560
    },
    {
      "entropy": 5.460619783401489,
      "epoch": 4.40538824383164,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003203169085662425,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21740400642156602,
      "num_tokens": 100809432.0,
      "step": 48565
    },
    {
      "entropy": 5.445725679397583,
      "epoch": 4.405841799709724,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003202852147471723,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2302703619003296,
      "num_tokens": 100819675.0,
      "step": 48570
    },
    {
      "entropy": 5.507003545761108,
      "epoch": 4.406295355587808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032025351999159237,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21817649900913239,
      "num_tokens": 100829779.0,
      "step": 48575
    },
    {
      "entropy": 5.472581911087036,
      "epoch": 4.406748911465892,
      "grad_norm": 0.9375,
      "learning_rate": 0.00032022182430015795,
      "loss": 4.9685,
      "mean_token_accuracy": 0.21350384652614593,
      "num_tokens": 100839840.0,
      "step": 48580
    },
    {
      "entropy": 5.430866384506226,
      "epoch": 4.407202467343977,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003201901276735248,
      "loss": 4.965,
      "mean_token_accuracy": 0.21884331852197647,
      "num_tokens": 100849495.0,
      "step": 48585
    },
    {
      "entropy": 5.46390290260315,
      "epoch": 4.407656023222061,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032015843011234826,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21769822537899017,
      "num_tokens": 100860214.0,
      "step": 48590
    },
    {
      "entropy": 5.6012670516967775,
      "epoch": 4.408109579100145,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032012673161728396,
      "loss": 5.0717,
      "mean_token_accuracy": 0.1996565505862236,
      "num_tokens": 100870773.0,
      "step": 48595
    },
    {
      "entropy": 5.53637957572937,
      "epoch": 4.4085631349782295,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032009503218898716,
      "loss": 4.9392,
      "mean_token_accuracy": 0.22835061103105544,
      "num_tokens": 100880654.0,
      "step": 48600
    },
    {
      "entropy": 5.401504039764404,
      "epoch": 4.409016690856314,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003200633318281138,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2162269338965416,
      "num_tokens": 100891217.0,
      "step": 48605
    },
    {
      "entropy": 5.417293310165405,
      "epoch": 4.409470246734398,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032003163053531913,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2235400691628456,
      "num_tokens": 100901424.0,
      "step": 48610
    },
    {
      "entropy": 5.510566854476929,
      "epoch": 4.409923802612482,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031999992831125893,
      "loss": 5.0307,
      "mean_token_accuracy": 0.21634821146726607,
      "num_tokens": 100912069.0,
      "step": 48615
    },
    {
      "entropy": 5.498762702941894,
      "epoch": 4.410377358490566,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003199682251565887,
      "loss": 4.913,
      "mean_token_accuracy": 0.2104746088385582,
      "num_tokens": 100922750.0,
      "step": 48620
    },
    {
      "entropy": 5.523212718963623,
      "epoch": 4.41083091436865,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031993652107196416,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2155785322189331,
      "num_tokens": 100933052.0,
      "step": 48625
    },
    {
      "entropy": 5.508798837661743,
      "epoch": 4.411284470246734,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031990481605804084,
      "loss": 4.9584,
      "mean_token_accuracy": 0.22102542817592621,
      "num_tokens": 100944519.0,
      "step": 48630
    },
    {
      "entropy": 5.44287519454956,
      "epoch": 4.4117380261248185,
      "grad_norm": 1.5625,
      "learning_rate": 0.00031987311011547455,
      "loss": 4.9584,
      "mean_token_accuracy": 0.21303386241197586,
      "num_tokens": 100954901.0,
      "step": 48635
    },
    {
      "entropy": 5.463715648651123,
      "epoch": 4.412191582002903,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003198414032449209,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22148942202329636,
      "num_tokens": 100964931.0,
      "step": 48640
    },
    {
      "entropy": 5.509176921844483,
      "epoch": 4.412645137880987,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003198096954470356,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22168587893247604,
      "num_tokens": 100974877.0,
      "step": 48645
    },
    {
      "entropy": 5.525095510482788,
      "epoch": 4.413098693759071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031977798672247444,
      "loss": 5.0257,
      "mean_token_accuracy": 0.2164207875728607,
      "num_tokens": 100986122.0,
      "step": 48650
    },
    {
      "entropy": 5.504905796051025,
      "epoch": 4.4135522496371555,
      "grad_norm": 0.984375,
      "learning_rate": 0.00031974627707189303,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21430962830781936,
      "num_tokens": 100996489.0,
      "step": 48655
    },
    {
      "entropy": 5.480649185180664,
      "epoch": 4.41400580551524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031971456649594723,
      "loss": 4.9883,
      "mean_token_accuracy": 0.21877104341983794,
      "num_tokens": 101006453.0,
      "step": 48660
    },
    {
      "entropy": 5.530335998535156,
      "epoch": 4.414459361393324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003196828549952927,
      "loss": 4.9826,
      "mean_token_accuracy": 0.21696620732545852,
      "num_tokens": 101015609.0,
      "step": 48665
    },
    {
      "entropy": 5.519145393371582,
      "epoch": 4.4149129172714074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003196511425705855,
      "loss": 4.9747,
      "mean_token_accuracy": 0.21820787340402603,
      "num_tokens": 101027406.0,
      "step": 48670
    },
    {
      "entropy": 5.430570840835571,
      "epoch": 4.415366473149492,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003196194292224812,
      "loss": 4.912,
      "mean_token_accuracy": 0.2142666444182396,
      "num_tokens": 101038255.0,
      "step": 48675
    },
    {
      "entropy": 5.436281013488769,
      "epoch": 4.415820029027576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031958771495163574,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21692990511655807,
      "num_tokens": 101048857.0,
      "step": 48680
    },
    {
      "entropy": 5.542431306838989,
      "epoch": 4.41627358490566,
      "grad_norm": 1.0625,
      "learning_rate": 0.000319555999758705,
      "loss": 4.9995,
      "mean_token_accuracy": 0.22021664083003997,
      "num_tokens": 101058640.0,
      "step": 48685
    },
    {
      "entropy": 5.552277135848999,
      "epoch": 4.4167271407837445,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031952428364434484,
      "loss": 5.0952,
      "mean_token_accuracy": 0.21393780112266542,
      "num_tokens": 101069537.0,
      "step": 48690
    },
    {
      "entropy": 5.509677171707153,
      "epoch": 4.417180696661829,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00031949256660921104,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20872405916452408,
      "num_tokens": 101080202.0,
      "step": 48695
    },
    {
      "entropy": 5.470819664001465,
      "epoch": 4.417634252539913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031946084865395975,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21651345044374465,
      "num_tokens": 101089500.0,
      "step": 48700
    },
    {
      "entropy": 5.568407440185547,
      "epoch": 4.418087808417997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031942912977924674,
      "loss": 5.0614,
      "mean_token_accuracy": 0.20611976832151413,
      "num_tokens": 101099858.0,
      "step": 48705
    },
    {
      "entropy": 5.510346794128418,
      "epoch": 4.418541364296082,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031939740998572806,
      "loss": 5.017,
      "mean_token_accuracy": 0.21115260124206542,
      "num_tokens": 101110834.0,
      "step": 48710
    },
    {
      "entropy": 5.528863000869751,
      "epoch": 4.418994920174166,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003193656892740595,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21567955762147903,
      "num_tokens": 101121003.0,
      "step": 48715
    },
    {
      "entropy": 5.566798067092895,
      "epoch": 4.419448476052249,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003193339676448973,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21447180062532425,
      "num_tokens": 101131925.0,
      "step": 48720
    },
    {
      "entropy": 5.468820190429687,
      "epoch": 4.4199020319303335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003193022450988973,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2224050924181938,
      "num_tokens": 101143130.0,
      "step": 48725
    },
    {
      "entropy": 5.4385514736175535,
      "epoch": 4.420355587808418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003192705216367156,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2223670780658722,
      "num_tokens": 101155524.0,
      "step": 48730
    },
    {
      "entropy": 5.428422260284424,
      "epoch": 4.420809143686502,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00031923879725900817,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22246858924627305,
      "num_tokens": 101164815.0,
      "step": 48735
    },
    {
      "entropy": 5.487641525268555,
      "epoch": 4.421262699564586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031920707196643125,
      "loss": 5.0302,
      "mean_token_accuracy": 0.21315535455942153,
      "num_tokens": 101174910.0,
      "step": 48740
    },
    {
      "entropy": 5.419308185577393,
      "epoch": 4.4217162554426706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003191753457596407,
      "loss": 4.8094,
      "mean_token_accuracy": 0.22575214356184006,
      "num_tokens": 101185698.0,
      "step": 48745
    },
    {
      "entropy": 5.461945533752441,
      "epoch": 4.422169811320755,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031914361863929276,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21456534564495086,
      "num_tokens": 101197001.0,
      "step": 48750
    },
    {
      "entropy": 5.437219142913818,
      "epoch": 4.422623367198839,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003191118906060436,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2196693241596222,
      "num_tokens": 101206705.0,
      "step": 48755
    },
    {
      "entropy": 5.480286598205566,
      "epoch": 4.423076923076923,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031908016166054933,
      "loss": 4.9407,
      "mean_token_accuracy": 0.2182798817753792,
      "num_tokens": 101216484.0,
      "step": 48760
    },
    {
      "entropy": 5.4818824291229244,
      "epoch": 4.423530478955008,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003190484318034659,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21482502222061156,
      "num_tokens": 101227169.0,
      "step": 48765
    },
    {
      "entropy": 5.497786664962769,
      "epoch": 4.423984034833092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003190167010354499,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22266385108232498,
      "num_tokens": 101236956.0,
      "step": 48770
    },
    {
      "entropy": 5.45860800743103,
      "epoch": 4.424437590711175,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003189849693571572,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21658395528793334,
      "num_tokens": 101248681.0,
      "step": 48775
    },
    {
      "entropy": 5.521940898895264,
      "epoch": 4.4248911465892595,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003189532367692441,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21687931418418885,
      "num_tokens": 101258375.0,
      "step": 48780
    },
    {
      "entropy": 5.4738609790802,
      "epoch": 4.425344702467344,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031892150327236693,
      "loss": 4.912,
      "mean_token_accuracy": 0.22437286674976348,
      "num_tokens": 101268513.0,
      "step": 48785
    },
    {
      "entropy": 5.495556020736695,
      "epoch": 4.425798258345428,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003188897688671818,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20982674211263658,
      "num_tokens": 101278998.0,
      "step": 48790
    },
    {
      "entropy": 5.507326936721801,
      "epoch": 4.426251814223512,
      "grad_norm": 0.921875,
      "learning_rate": 0.00031885803355434516,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21697127521038057,
      "num_tokens": 101289090.0,
      "step": 48795
    },
    {
      "entropy": 5.492192649841309,
      "epoch": 4.426705370101597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003188262973345131,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21309439092874527,
      "num_tokens": 101299697.0,
      "step": 48800
    },
    {
      "entropy": 5.525406265258789,
      "epoch": 4.427158925979681,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031879456020834213,
      "loss": 5.0006,
      "mean_token_accuracy": 0.21023459881544113,
      "num_tokens": 101310470.0,
      "step": 48805
    },
    {
      "entropy": 5.513312244415284,
      "epoch": 4.427612481857765,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031876282217648845,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21869086772203444,
      "num_tokens": 101321864.0,
      "step": 48810
    },
    {
      "entropy": 5.485288095474243,
      "epoch": 4.428066037735849,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031873108323960845,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2202020674943924,
      "num_tokens": 101331302.0,
      "step": 48815
    },
    {
      "entropy": 5.472030401229858,
      "epoch": 4.428519593613933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031869934339835856,
      "loss": 4.9977,
      "mean_token_accuracy": 0.21406537145376206,
      "num_tokens": 101341387.0,
      "step": 48820
    },
    {
      "entropy": 5.497788190841675,
      "epoch": 4.428973149492017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031866760265339514,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22022034376859664,
      "num_tokens": 101351235.0,
      "step": 48825
    },
    {
      "entropy": 5.468705177307129,
      "epoch": 4.429426705370101,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031863586100537453,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21400896608829498,
      "num_tokens": 101361289.0,
      "step": 48830
    },
    {
      "entropy": 5.427786779403687,
      "epoch": 4.429880261248186,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003186041184549531,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21966639757156373,
      "num_tokens": 101371666.0,
      "step": 48835
    },
    {
      "entropy": 5.455482816696167,
      "epoch": 4.43033381712627,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003185723750027875,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22375288903713225,
      "num_tokens": 101381195.0,
      "step": 48840
    },
    {
      "entropy": 5.576057815551758,
      "epoch": 4.430787373004354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031854063064953403,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21556016504764558,
      "num_tokens": 101392478.0,
      "step": 48845
    },
    {
      "entropy": 5.553357219696045,
      "epoch": 4.431240928882438,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003185088853958492,
      "loss": 5.0281,
      "mean_token_accuracy": 0.2113853096961975,
      "num_tokens": 101405005.0,
      "step": 48850
    },
    {
      "entropy": 5.471109867095947,
      "epoch": 4.431694484760523,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031847713924238965,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21333996653556825,
      "num_tokens": 101416588.0,
      "step": 48855
    },
    {
      "entropy": 5.488896989822388,
      "epoch": 4.432148040638607,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031844539218981165,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2141649082303047,
      "num_tokens": 101426784.0,
      "step": 48860
    },
    {
      "entropy": 5.440081262588501,
      "epoch": 4.432601596516691,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003184136442387719,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21638728380203248,
      "num_tokens": 101437919.0,
      "step": 48865
    },
    {
      "entropy": 5.473404455184936,
      "epoch": 4.433055152394775,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031838189538992693,
      "loss": 4.9877,
      "mean_token_accuracy": 0.21494482904672624,
      "num_tokens": 101448654.0,
      "step": 48870
    },
    {
      "entropy": 5.489623260498047,
      "epoch": 4.433508708272859,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003183501456439334,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22196384966373445,
      "num_tokens": 101458648.0,
      "step": 48875
    },
    {
      "entropy": 5.453811550140381,
      "epoch": 4.433962264150943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031831839500144767,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21487681120634078,
      "num_tokens": 101469339.0,
      "step": 48880
    },
    {
      "entropy": 5.487022447586059,
      "epoch": 4.434415820029027,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031828664346312657,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2201332688331604,
      "num_tokens": 101479452.0,
      "step": 48885
    },
    {
      "entropy": 5.459632349014282,
      "epoch": 4.434869375907112,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003182548910296267,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2172624424099922,
      "num_tokens": 101489147.0,
      "step": 48890
    },
    {
      "entropy": 5.433256483078003,
      "epoch": 4.435322931785196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003182231377016046,
      "loss": 4.8418,
      "mean_token_accuracy": 0.22489100992679595,
      "num_tokens": 101498958.0,
      "step": 48895
    },
    {
      "entropy": 5.510187244415283,
      "epoch": 4.43577648766328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000318191383479717,
      "loss": 5.0143,
      "mean_token_accuracy": 0.21233898550271987,
      "num_tokens": 101509197.0,
      "step": 48900
    },
    {
      "entropy": 5.517392349243164,
      "epoch": 4.436230043541364,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003181596283646205,
      "loss": 4.966,
      "mean_token_accuracy": 0.21737496852874755,
      "num_tokens": 101520018.0,
      "step": 48905
    },
    {
      "entropy": 5.482361507415772,
      "epoch": 4.436683599419449,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00031812787235697206,
      "loss": 4.9484,
      "mean_token_accuracy": 0.21748346239328384,
      "num_tokens": 101530755.0,
      "step": 48910
    },
    {
      "entropy": 5.443240451812744,
      "epoch": 4.437137155297533,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031809611545742804,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20898209065198897,
      "num_tokens": 101541028.0,
      "step": 48915
    },
    {
      "entropy": 5.487020349502563,
      "epoch": 4.437590711175617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003180643576666455,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22013065367937087,
      "num_tokens": 101550826.0,
      "step": 48920
    },
    {
      "entropy": 5.542946720123291,
      "epoch": 4.4380442670537015,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003180325989852811,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2186713233590126,
      "num_tokens": 101560657.0,
      "step": 48925
    },
    {
      "entropy": 5.446054267883301,
      "epoch": 4.438497822931785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003180008394139914,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2182600900530815,
      "num_tokens": 101570504.0,
      "step": 48930
    },
    {
      "entropy": 5.508131885528565,
      "epoch": 4.438951378809869,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031796907895343356,
      "loss": 5.0488,
      "mean_token_accuracy": 0.20786026269197463,
      "num_tokens": 101581729.0,
      "step": 48935
    },
    {
      "entropy": 5.564155530929566,
      "epoch": 4.439404934687953,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00031793731760426414,
      "loss": 5.0383,
      "mean_token_accuracy": 0.21208385229110718,
      "num_tokens": 101592649.0,
      "step": 48940
    },
    {
      "entropy": 5.581769418716431,
      "epoch": 4.439858490566038,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031790555536714013,
      "loss": 5.1077,
      "mean_token_accuracy": 0.20836289823055268,
      "num_tokens": 101603824.0,
      "step": 48945
    },
    {
      "entropy": 5.498068237304688,
      "epoch": 4.440312046444122,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003178737922427182,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21887848973274232,
      "num_tokens": 101615659.0,
      "step": 48950
    },
    {
      "entropy": 5.567192554473877,
      "epoch": 4.440765602322206,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003178420282316554,
      "loss": 5.0383,
      "mean_token_accuracy": 0.2103719025850296,
      "num_tokens": 101625937.0,
      "step": 48955
    },
    {
      "entropy": 5.507747793197632,
      "epoch": 4.4412191582002905,
      "grad_norm": 1.0,
      "learning_rate": 0.0003178102633346085,
      "loss": 4.9088,
      "mean_token_accuracy": 0.22426542192697524,
      "num_tokens": 101637352.0,
      "step": 48960
    },
    {
      "entropy": 5.495343828201294,
      "epoch": 4.441672714078375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003177784975522344,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22541967928409576,
      "num_tokens": 101647939.0,
      "step": 48965
    },
    {
      "entropy": 5.527831363677978,
      "epoch": 4.442126269956459,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031774673088519014,
      "loss": 4.9997,
      "mean_token_accuracy": 0.2091417580842972,
      "num_tokens": 101658200.0,
      "step": 48970
    },
    {
      "entropy": 5.5058294296264645,
      "epoch": 4.442579825834542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003177149633341326,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21151622384786606,
      "num_tokens": 101668624.0,
      "step": 48975
    },
    {
      "entropy": 5.527393341064453,
      "epoch": 4.443033381712627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003176831948997187,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20939749032258986,
      "num_tokens": 101678939.0,
      "step": 48980
    },
    {
      "entropy": 5.498007440567017,
      "epoch": 4.443486937590711,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003176514255826054,
      "loss": 4.9901,
      "mean_token_accuracy": 0.22075887769460678,
      "num_tokens": 101689365.0,
      "step": 48985
    },
    {
      "entropy": 5.448480415344238,
      "epoch": 4.443940493468795,
      "grad_norm": 1.5,
      "learning_rate": 0.0003176196553834498,
      "loss": 4.923,
      "mean_token_accuracy": 0.21172663420438767,
      "num_tokens": 101699429.0,
      "step": 48990
    },
    {
      "entropy": 5.494911813735962,
      "epoch": 4.444394049346879,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031758788430290893,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2152695581316948,
      "num_tokens": 101709452.0,
      "step": 48995
    },
    {
      "entropy": 5.487617874145508,
      "epoch": 4.444847605224964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031755611234163966,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2215028077363968,
      "num_tokens": 101719783.0,
      "step": 49000
    },
    {
      "entropy": 5.455603551864624,
      "epoch": 4.445301161103048,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003175243395002992,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22631808668375014,
      "num_tokens": 101729100.0,
      "step": 49005
    },
    {
      "entropy": 5.491920042037964,
      "epoch": 4.445754716981132,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031749256577954453,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21982122957706451,
      "num_tokens": 101738711.0,
      "step": 49010
    },
    {
      "entropy": 5.447384214401245,
      "epoch": 4.4462082728592165,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003174607911800328,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21971216201782226,
      "num_tokens": 101749624.0,
      "step": 49015
    },
    {
      "entropy": 5.440544939041137,
      "epoch": 4.446661828737301,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00031742901570242107,
      "loss": 4.897,
      "mean_token_accuracy": 0.21929822266101837,
      "num_tokens": 101760951.0,
      "step": 49020
    },
    {
      "entropy": 5.468209648132325,
      "epoch": 4.447115384615385,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031739723934736655,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22215349674224855,
      "num_tokens": 101770773.0,
      "step": 49025
    },
    {
      "entropy": 5.506128597259521,
      "epoch": 4.447568940493468,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003173654621155262,
      "loss": 4.9798,
      "mean_token_accuracy": 0.2129996418952942,
      "num_tokens": 101781566.0,
      "step": 49030
    },
    {
      "entropy": 5.419765472412109,
      "epoch": 4.448022496371553,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031733368400755735,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22364138066768646,
      "num_tokens": 101792268.0,
      "step": 49035
    },
    {
      "entropy": 5.5335602283477785,
      "epoch": 4.448476052249637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031730190502411715,
      "loss": 4.9999,
      "mean_token_accuracy": 0.21340702027082442,
      "num_tokens": 101802500.0,
      "step": 49040
    },
    {
      "entropy": 5.5416820526123045,
      "epoch": 4.448929608127721,
      "grad_norm": 1.0,
      "learning_rate": 0.0003172701251658628,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21748762726783752,
      "num_tokens": 101813851.0,
      "step": 49045
    },
    {
      "entropy": 5.5869354724884035,
      "epoch": 4.4493831640058055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031723834443345134,
      "loss": 4.9894,
      "mean_token_accuracy": 0.21207403689622878,
      "num_tokens": 101824774.0,
      "step": 49050
    },
    {
      "entropy": 5.416728687286377,
      "epoch": 4.44983671988389,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031720656282754027,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23302250504493713,
      "num_tokens": 101835975.0,
      "step": 49055
    },
    {
      "entropy": 5.405594730377198,
      "epoch": 4.450290275761974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031717478034878676,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22239863127470016,
      "num_tokens": 101846764.0,
      "step": 49060
    },
    {
      "entropy": 5.448556280136108,
      "epoch": 4.450743831640058,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031714299699784796,
      "loss": 4.9466,
      "mean_token_accuracy": 0.22329875379800795,
      "num_tokens": 101857968.0,
      "step": 49065
    },
    {
      "entropy": 5.479585647583008,
      "epoch": 4.4511973875181425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003171112127753813,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22911406904459,
      "num_tokens": 101868484.0,
      "step": 49070
    },
    {
      "entropy": 5.4896996974945065,
      "epoch": 4.451650943396227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003170794276820441,
      "loss": 4.947,
      "mean_token_accuracy": 0.2188268408179283,
      "num_tokens": 101877969.0,
      "step": 49075
    },
    {
      "entropy": 5.441675090789795,
      "epoch": 4.452104499274311,
      "grad_norm": 1.0,
      "learning_rate": 0.00031704764171849344,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2208358734846115,
      "num_tokens": 101888635.0,
      "step": 49080
    },
    {
      "entropy": 5.472956037521362,
      "epoch": 4.4525580551523944,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031701585488538695,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22742778807878494,
      "num_tokens": 101898340.0,
      "step": 49085
    },
    {
      "entropy": 5.5568015575408936,
      "epoch": 4.453011611030479,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003169840671833819,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20809100866317748,
      "num_tokens": 101908993.0,
      "step": 49090
    },
    {
      "entropy": 5.450288867950439,
      "epoch": 4.453465166908563,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003169522786131355,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21743710935115815,
      "num_tokens": 101919675.0,
      "step": 49095
    },
    {
      "entropy": 5.534206485748291,
      "epoch": 4.453918722786647,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031692048917530535,
      "loss": 5.0168,
      "mean_token_accuracy": 0.21145865470170974,
      "num_tokens": 101929949.0,
      "step": 49100
    },
    {
      "entropy": 5.453121852874756,
      "epoch": 4.4543722786647315,
      "grad_norm": 1.125,
      "learning_rate": 0.00031688869887054887,
      "loss": 4.8138,
      "mean_token_accuracy": 0.2343791678547859,
      "num_tokens": 101940025.0,
      "step": 49105
    },
    {
      "entropy": 5.455269861221313,
      "epoch": 4.454825834542816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031685690769952336,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22717441171407698,
      "num_tokens": 101950344.0,
      "step": 49110
    },
    {
      "entropy": 5.5091125011444095,
      "epoch": 4.4552793904209,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003168251156628863,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2150905191898346,
      "num_tokens": 101961405.0,
      "step": 49115
    },
    {
      "entropy": 5.3848518371582035,
      "epoch": 4.455732946298984,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031679332276129523,
      "loss": 4.8136,
      "mean_token_accuracy": 0.22729538679122924,
      "num_tokens": 101972164.0,
      "step": 49120
    },
    {
      "entropy": 5.439436960220337,
      "epoch": 4.456186502177069,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031676152899540763,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2141142651438713,
      "num_tokens": 101981573.0,
      "step": 49125
    },
    {
      "entropy": 5.47204704284668,
      "epoch": 4.456640058055152,
      "grad_norm": 1.0625,
      "learning_rate": 0.000316729734365881,
      "loss": 4.9046,
      "mean_token_accuracy": 0.21977248042821884,
      "num_tokens": 101991051.0,
      "step": 49130
    },
    {
      "entropy": 5.477607297897339,
      "epoch": 4.457093613933236,
      "grad_norm": 1.0,
      "learning_rate": 0.0003166979388733727,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22396069765090942,
      "num_tokens": 102000003.0,
      "step": 49135
    },
    {
      "entropy": 5.48523211479187,
      "epoch": 4.4575471698113205,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003166661425185405,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2105197548866272,
      "num_tokens": 102010489.0,
      "step": 49140
    },
    {
      "entropy": 5.4840929985046385,
      "epoch": 4.458000725689405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031663434530204183,
      "loss": 5.0017,
      "mean_token_accuracy": 0.20485190004110337,
      "num_tokens": 102020486.0,
      "step": 49145
    },
    {
      "entropy": 5.509133386611938,
      "epoch": 4.458454281567489,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003166025472245342,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21942233592271804,
      "num_tokens": 102030098.0,
      "step": 49150
    },
    {
      "entropy": 5.519683074951172,
      "epoch": 4.458907837445573,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003165707482866754,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21721022725105285,
      "num_tokens": 102040850.0,
      "step": 49155
    },
    {
      "entropy": 5.432465887069702,
      "epoch": 4.4593613933236576,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031653894848912296,
      "loss": 4.87,
      "mean_token_accuracy": 0.2204096645116806,
      "num_tokens": 102051056.0,
      "step": 49160
    },
    {
      "entropy": 5.509150505065918,
      "epoch": 4.459814949201742,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003165071478325344,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21620594412088395,
      "num_tokens": 102061727.0,
      "step": 49165
    },
    {
      "entropy": 5.472674942016601,
      "epoch": 4.460268505079826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003164753463175675,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21426450461149216,
      "num_tokens": 102071501.0,
      "step": 49170
    },
    {
      "entropy": 5.502066040039063,
      "epoch": 4.46072206095791,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031644354394487977,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2256367325782776,
      "num_tokens": 102082511.0,
      "step": 49175
    },
    {
      "entropy": 5.513202142715454,
      "epoch": 4.461175616835995,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031641174071512905,
      "loss": 4.9468,
      "mean_token_accuracy": 0.2265630140900612,
      "num_tokens": 102093010.0,
      "step": 49180
    },
    {
      "entropy": 5.456246471405029,
      "epoch": 4.461629172714078,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000316379936628973,
      "loss": 4.9304,
      "mean_token_accuracy": 0.2206541508436203,
      "num_tokens": 102102691.0,
      "step": 49185
    },
    {
      "entropy": 5.502698850631714,
      "epoch": 4.462082728592162,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031634813168706933,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2219164714217186,
      "num_tokens": 102112726.0,
      "step": 49190
    },
    {
      "entropy": 5.461038446426391,
      "epoch": 4.4625362844702465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031631632589007575,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2216406136751175,
      "num_tokens": 102122879.0,
      "step": 49195
    },
    {
      "entropy": 5.50758810043335,
      "epoch": 4.462989840348331,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031628451923864997,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21900622099637984,
      "num_tokens": 102133485.0,
      "step": 49200
    },
    {
      "entropy": 5.52626223564148,
      "epoch": 4.463443396226415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031625271173344984,
      "loss": 5.0139,
      "mean_token_accuracy": 0.2133538618683815,
      "num_tokens": 102143439.0,
      "step": 49205
    },
    {
      "entropy": 5.437586641311645,
      "epoch": 4.463896952104499,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031622090337513314,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22447779923677444,
      "num_tokens": 102153808.0,
      "step": 49210
    },
    {
      "entropy": 5.562706422805786,
      "epoch": 4.464350507982584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031618909416435766,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20915221869945527,
      "num_tokens": 102163375.0,
      "step": 49215
    },
    {
      "entropy": 5.530718374252319,
      "epoch": 4.464804063860668,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003161572841017811,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2162939727306366,
      "num_tokens": 102174402.0,
      "step": 49220
    },
    {
      "entropy": 5.416842412948609,
      "epoch": 4.465257619738752,
      "grad_norm": 1.0,
      "learning_rate": 0.00031612547318806154,
      "loss": 4.7972,
      "mean_token_accuracy": 0.22899212092161178,
      "num_tokens": 102184446.0,
      "step": 49225
    },
    {
      "entropy": 5.460968065261841,
      "epoch": 4.465711175616836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031609366142385675,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21674534678459167,
      "num_tokens": 102194398.0,
      "step": 49230
    },
    {
      "entropy": 5.402218627929687,
      "epoch": 4.46616473149492,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003160618488098245,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22297408878803254,
      "num_tokens": 102205953.0,
      "step": 49235
    },
    {
      "entropy": 5.450443363189697,
      "epoch": 4.466618287373004,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031603003534662273,
      "loss": 4.8972,
      "mean_token_accuracy": 0.2258635252714157,
      "num_tokens": 102215815.0,
      "step": 49240
    },
    {
      "entropy": 5.4714953899383545,
      "epoch": 4.467071843251088,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00031599822103490946,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22198066711425782,
      "num_tokens": 102225703.0,
      "step": 49245
    },
    {
      "entropy": 5.555132484436035,
      "epoch": 4.467525399129173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003159664058753425,
      "loss": 5.0704,
      "mean_token_accuracy": 0.20212515741586684,
      "num_tokens": 102236627.0,
      "step": 49250
    },
    {
      "entropy": 5.488466548919678,
      "epoch": 4.467978955007257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003159345898685797,
      "loss": 4.964,
      "mean_token_accuracy": 0.22260389924049379,
      "num_tokens": 102246539.0,
      "step": 49255
    },
    {
      "entropy": 5.475691127777099,
      "epoch": 4.468432510885341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003159027730152793,
      "loss": 4.9615,
      "mean_token_accuracy": 0.219437775015831,
      "num_tokens": 102256734.0,
      "step": 49260
    },
    {
      "entropy": 5.5335053443908695,
      "epoch": 4.468886066763425,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003158709553160991,
      "loss": 5.0745,
      "mean_token_accuracy": 0.20860497206449508,
      "num_tokens": 102267469.0,
      "step": 49265
    },
    {
      "entropy": 5.526206111907959,
      "epoch": 4.46933962264151,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00031583913677169705,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21543705463409424,
      "num_tokens": 102278435.0,
      "step": 49270
    },
    {
      "entropy": 5.579649496078491,
      "epoch": 4.469793178519594,
      "grad_norm": 1.125,
      "learning_rate": 0.00031580731738273135,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21661664098501204,
      "num_tokens": 102289171.0,
      "step": 49275
    },
    {
      "entropy": 5.4825828075408936,
      "epoch": 4.470246734397678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003157754971498599,
      "loss": 4.9079,
      "mean_token_accuracy": 0.23040841817855834,
      "num_tokens": 102299254.0,
      "step": 49280
    },
    {
      "entropy": 5.521515321731568,
      "epoch": 4.4707002902757615,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003157436760737408,
      "loss": 4.977,
      "mean_token_accuracy": 0.21250390708446504,
      "num_tokens": 102308471.0,
      "step": 49285
    },
    {
      "entropy": 5.504034471511841,
      "epoch": 4.471153846153846,
      "grad_norm": 1.09375,
      "learning_rate": 0.000315711854155032,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21520009487867356,
      "num_tokens": 102319005.0,
      "step": 49290
    },
    {
      "entropy": 5.473904371261597,
      "epoch": 4.47160740203193,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031568003139439185,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22264574319124222,
      "num_tokens": 102329568.0,
      "step": 49295
    },
    {
      "entropy": 5.58042950630188,
      "epoch": 4.472060957910014,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003156482077924781,
      "loss": 5.0647,
      "mean_token_accuracy": 0.21246519088745117,
      "num_tokens": 102340810.0,
      "step": 49300
    },
    {
      "entropy": 5.438494634628296,
      "epoch": 4.472514513788099,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003156163833499491,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21505184322595597,
      "num_tokens": 102351305.0,
      "step": 49305
    },
    {
      "entropy": 5.522437810897827,
      "epoch": 4.472968069666183,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031558455806746306,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21374068707227706,
      "num_tokens": 102362057.0,
      "step": 49310
    },
    {
      "entropy": 5.55515718460083,
      "epoch": 4.473421625544267,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003155527319456779,
      "loss": 5.089,
      "mean_token_accuracy": 0.20650786459445952,
      "num_tokens": 102372612.0,
      "step": 49315
    },
    {
      "entropy": 5.465106773376465,
      "epoch": 4.473875181422351,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00031552090498525195,
      "loss": 4.92,
      "mean_token_accuracy": 0.22260379046201706,
      "num_tokens": 102383337.0,
      "step": 49320
    },
    {
      "entropy": 5.525664138793945,
      "epoch": 4.474328737300436,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003154890771868433,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21994102150201797,
      "num_tokens": 102394196.0,
      "step": 49325
    },
    {
      "entropy": 5.558775949478149,
      "epoch": 4.47478229317852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031545724855111025,
      "loss": 4.9634,
      "mean_token_accuracy": 0.22246255576610566,
      "num_tokens": 102404677.0,
      "step": 49330
    },
    {
      "entropy": 5.471353340148926,
      "epoch": 4.475235849056604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031542541907871095,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22821212708950042,
      "num_tokens": 102413445.0,
      "step": 49335
    },
    {
      "entropy": 5.432743215560913,
      "epoch": 4.475689404934688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003153935887703038,
      "loss": 4.9453,
      "mean_token_accuracy": 0.2226989597082138,
      "num_tokens": 102424527.0,
      "step": 49340
    },
    {
      "entropy": 5.558220720291137,
      "epoch": 4.476142960812772,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003153617576265469,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2150629535317421,
      "num_tokens": 102434940.0,
      "step": 49345
    },
    {
      "entropy": 5.581736993789673,
      "epoch": 4.476596516690856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031532992564809847,
      "loss": 5.0239,
      "mean_token_accuracy": 0.2211265206336975,
      "num_tokens": 102446678.0,
      "step": 49350
    },
    {
      "entropy": 5.434978389739991,
      "epoch": 4.47705007256894,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000315298092835617,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22333578616380692,
      "num_tokens": 102457255.0,
      "step": 49355
    },
    {
      "entropy": 5.539587211608887,
      "epoch": 4.477503628447025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031526625918976066,
      "loss": 5.0742,
      "mean_token_accuracy": 0.21733324229717255,
      "num_tokens": 102468385.0,
      "step": 49360
    },
    {
      "entropy": 5.513609743118286,
      "epoch": 4.477957184325109,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031523442471118783,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21884564459323883,
      "num_tokens": 102478204.0,
      "step": 49365
    },
    {
      "entropy": 5.534934043884277,
      "epoch": 4.478410740203193,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031520258940055684,
      "loss": 4.9719,
      "mean_token_accuracy": 0.21817899048328399,
      "num_tokens": 102488492.0,
      "step": 49370
    },
    {
      "entropy": 5.4786028385162355,
      "epoch": 4.4788642960812775,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031517075325852606,
      "loss": 4.935,
      "mean_token_accuracy": 0.21973239481449128,
      "num_tokens": 102500309.0,
      "step": 49375
    },
    {
      "entropy": 5.452363538742065,
      "epoch": 4.479317851959362,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003151389162857539,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2187344342470169,
      "num_tokens": 102511196.0,
      "step": 49380
    },
    {
      "entropy": 5.456980133056641,
      "epoch": 4.479771407837446,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00031510707848289864,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2211582615971565,
      "num_tokens": 102522100.0,
      "step": 49385
    },
    {
      "entropy": 5.468001031875611,
      "epoch": 4.480224963715529,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031507523985061886,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2311003550887108,
      "num_tokens": 102531677.0,
      "step": 49390
    },
    {
      "entropy": 5.441324090957641,
      "epoch": 4.480678519593614,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031504340038957294,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23381798416376115,
      "num_tokens": 102541854.0,
      "step": 49395
    },
    {
      "entropy": 5.493488121032715,
      "epoch": 4.481132075471698,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003150115601004192,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21838894188404084,
      "num_tokens": 102551187.0,
      "step": 49400
    },
    {
      "entropy": 5.515256738662719,
      "epoch": 4.481585631349782,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003149797189838162,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21642646342515945,
      "num_tokens": 102562568.0,
      "step": 49405
    },
    {
      "entropy": 5.526375102996826,
      "epoch": 4.482039187227866,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031494787704042243,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21078622043132783,
      "num_tokens": 102573170.0,
      "step": 49410
    },
    {
      "entropy": 5.421289157867432,
      "epoch": 4.482492743105951,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031491603427089644,
      "loss": 4.7991,
      "mean_token_accuracy": 0.22796399146318436,
      "num_tokens": 102583950.0,
      "step": 49415
    },
    {
      "entropy": 5.474205064773559,
      "epoch": 4.482946298984035,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031488419067589665,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2165321797132492,
      "num_tokens": 102594370.0,
      "step": 49420
    },
    {
      "entropy": 5.517958402633667,
      "epoch": 4.483399854862119,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003148523462560816,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2201106995344162,
      "num_tokens": 102604837.0,
      "step": 49425
    },
    {
      "entropy": 5.502716541290283,
      "epoch": 4.4838534107402035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003148205010121099,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2158895418047905,
      "num_tokens": 102616170.0,
      "step": 49430
    },
    {
      "entropy": 5.480481481552124,
      "epoch": 4.484306966618288,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003147886549446401,
      "loss": 4.9424,
      "mean_token_accuracy": 0.2226933106780052,
      "num_tokens": 102627283.0,
      "step": 49435
    },
    {
      "entropy": 5.5754491806030275,
      "epoch": 4.484760522496371,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003147568080543308,
      "loss": 4.9719,
      "mean_token_accuracy": 0.2095937341451645,
      "num_tokens": 102637087.0,
      "step": 49440
    },
    {
      "entropy": 5.503161382675171,
      "epoch": 4.485214078374455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003147249603418406,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2191644787788391,
      "num_tokens": 102646818.0,
      "step": 49445
    },
    {
      "entropy": 5.397304344177246,
      "epoch": 4.48566763425254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003146931118078281,
      "loss": 4.8688,
      "mean_token_accuracy": 0.2305024415254593,
      "num_tokens": 102656940.0,
      "step": 49450
    },
    {
      "entropy": 5.433300352096557,
      "epoch": 4.486121190130624,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003146612624529518,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22086893618106843,
      "num_tokens": 102666841.0,
      "step": 49455
    },
    {
      "entropy": 5.524063682556152,
      "epoch": 4.486574746008708,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031462941227787056,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21305639147758484,
      "num_tokens": 102676821.0,
      "step": 49460
    },
    {
      "entropy": 5.607551145553589,
      "epoch": 4.4870283018867925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145975612832431,
      "loss": 5.0479,
      "mean_token_accuracy": 0.21334024518728256,
      "num_tokens": 102687340.0,
      "step": 49465
    },
    {
      "entropy": 5.469859504699707,
      "epoch": 4.487481857764877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003145657094697278,
      "loss": 4.9444,
      "mean_token_accuracy": 0.22017977982759476,
      "num_tokens": 102698320.0,
      "step": 49470
    },
    {
      "entropy": 5.543057298660278,
      "epoch": 4.487935413642961,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003145338568379836,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21705076694488526,
      "num_tokens": 102708685.0,
      "step": 49475
    },
    {
      "entropy": 5.471353197097779,
      "epoch": 4.488388969521045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003145020033886692,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2128371000289917,
      "num_tokens": 102719680.0,
      "step": 49480
    },
    {
      "entropy": 5.4166923522949215,
      "epoch": 4.4888425253991295,
      "grad_norm": 0.984375,
      "learning_rate": 0.00031447014912244325,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2224338635802269,
      "num_tokens": 102731156.0,
      "step": 49485
    },
    {
      "entropy": 5.470530939102173,
      "epoch": 4.489296081277214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003144382940399646,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22455713748931885,
      "num_tokens": 102740790.0,
      "step": 49490
    },
    {
      "entropy": 5.484751129150391,
      "epoch": 4.489749637155297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000314406438141892,
      "loss": 4.9288,
      "mean_token_accuracy": 0.22127767354249955,
      "num_tokens": 102750672.0,
      "step": 49495
    },
    {
      "entropy": 5.415508365631103,
      "epoch": 4.4902031930333814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003143745814288842,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2117939203977585,
      "num_tokens": 102760364.0,
      "step": 49500
    },
    {
      "entropy": 5.581341934204102,
      "epoch": 4.490656748911466,
      "grad_norm": 1.0,
      "learning_rate": 0.0003143427239016001,
      "loss": 5.0301,
      "mean_token_accuracy": 0.21565532982349395,
      "num_tokens": 102770415.0,
      "step": 49505
    },
    {
      "entropy": 5.527942085266114,
      "epoch": 4.49111030478955,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003143108655606983,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2253287062048912,
      "num_tokens": 102781203.0,
      "step": 49510
    },
    {
      "entropy": 5.46514663696289,
      "epoch": 4.491563860667634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031427900640683783,
      "loss": 4.894,
      "mean_token_accuracy": 0.22216376662254333,
      "num_tokens": 102791071.0,
      "step": 49515
    },
    {
      "entropy": 5.506223154067993,
      "epoch": 4.4920174165457185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031424714644067754,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20806125700473785,
      "num_tokens": 102801755.0,
      "step": 49520
    },
    {
      "entropy": 5.512821626663208,
      "epoch": 4.492470972423803,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031421528566287623,
      "loss": 4.98,
      "mean_token_accuracy": 0.21211925595998765,
      "num_tokens": 102812947.0,
      "step": 49525
    },
    {
      "entropy": 5.4641670227050785,
      "epoch": 4.492924528301887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003141834240740928,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2229432612657547,
      "num_tokens": 102822917.0,
      "step": 49530
    },
    {
      "entropy": 5.488776159286499,
      "epoch": 4.493378084179971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003141515616749862,
      "loss": 4.9213,
      "mean_token_accuracy": 0.2173803150653839,
      "num_tokens": 102833102.0,
      "step": 49535
    },
    {
      "entropy": 5.556230115890503,
      "epoch": 4.493831640058055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003141196984662153,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2222593292593956,
      "num_tokens": 102843888.0,
      "step": 49540
    },
    {
      "entropy": 5.531874322891236,
      "epoch": 4.494285195936139,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003140878344484391,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2206108257174492,
      "num_tokens": 102854417.0,
      "step": 49545
    },
    {
      "entropy": 5.4544330596923825,
      "epoch": 4.494738751814223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003140559696223165,
      "loss": 4.9221,
      "mean_token_accuracy": 0.2236522376537323,
      "num_tokens": 102865206.0,
      "step": 49550
    },
    {
      "entropy": 5.503578567504883,
      "epoch": 4.4951923076923075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003140241039885065,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2225091576576233,
      "num_tokens": 102874967.0,
      "step": 49555
    },
    {
      "entropy": 5.440956544876099,
      "epoch": 4.495645863570392,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003139922375476681,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22025995552539826,
      "num_tokens": 102884808.0,
      "step": 49560
    },
    {
      "entropy": 5.407374429702759,
      "epoch": 4.496099419448476,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003139603703004603,
      "loss": 4.8284,
      "mean_token_accuracy": 0.22469656318426132,
      "num_tokens": 102895206.0,
      "step": 49565
    },
    {
      "entropy": 5.514058399200439,
      "epoch": 4.49655297532656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003139285022475421,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21810391098260878,
      "num_tokens": 102906658.0,
      "step": 49570
    },
    {
      "entropy": 5.45593490600586,
      "epoch": 4.4970065312046446,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031389663338957254,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2186199426651001,
      "num_tokens": 102917071.0,
      "step": 49575
    },
    {
      "entropy": 5.440912199020386,
      "epoch": 4.497460087082729,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031386476372721067,
      "loss": 4.887,
      "mean_token_accuracy": 0.22219536155462266,
      "num_tokens": 102927280.0,
      "step": 49580
    },
    {
      "entropy": 5.514660453796386,
      "epoch": 4.497913642960813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003138328932611156,
      "loss": 5.0047,
      "mean_token_accuracy": 0.21142062395811081,
      "num_tokens": 102937670.0,
      "step": 49585
    },
    {
      "entropy": 5.507161855697632,
      "epoch": 4.498367198838897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031380102199194646,
      "loss": 4.9574,
      "mean_token_accuracy": 0.217823788523674,
      "num_tokens": 102947874.0,
      "step": 49590
    },
    {
      "entropy": 5.552669954299927,
      "epoch": 4.498820754716981,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003137691499203622,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2111913189291954,
      "num_tokens": 102959311.0,
      "step": 49595
    },
    {
      "entropy": 5.4854652881622314,
      "epoch": 4.499274310595065,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031373727704702217,
      "loss": 4.9068,
      "mean_token_accuracy": 0.21774058192968368,
      "num_tokens": 102969866.0,
      "step": 49600
    },
    {
      "entropy": 5.444125318527222,
      "epoch": 4.499727866473149,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031370540337258523,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22624369114637374,
      "num_tokens": 102979672.0,
      "step": 49605
    },
    {
      "entropy": 5.450902891159058,
      "epoch": 4.5001814223512335,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003136735288977108,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21696875542402266,
      "num_tokens": 102991050.0,
      "step": 49610
    },
    {
      "entropy": 5.492629480361939,
      "epoch": 4.500634978229318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031364165362305783,
      "loss": 4.979,
      "mean_token_accuracy": 0.21614715158939363,
      "num_tokens": 103001700.0,
      "step": 49615
    },
    {
      "entropy": 5.536313247680664,
      "epoch": 4.501088534107402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003136097775492857,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21474287509918213,
      "num_tokens": 103013013.0,
      "step": 49620
    },
    {
      "entropy": 5.48390941619873,
      "epoch": 4.501542089985486,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003135779006770535,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21923365741968154,
      "num_tokens": 103024110.0,
      "step": 49625
    },
    {
      "entropy": 5.555691480636597,
      "epoch": 4.501995645863571,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003135460230070205,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21541413366794587,
      "num_tokens": 103034991.0,
      "step": 49630
    },
    {
      "entropy": 5.527625846862793,
      "epoch": 4.502449201741655,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031351414453984595,
      "loss": 4.9707,
      "mean_token_accuracy": 0.217935049533844,
      "num_tokens": 103045634.0,
      "step": 49635
    },
    {
      "entropy": 5.446816539764404,
      "epoch": 4.502902757619739,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000313482265276189,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21912679672241211,
      "num_tokens": 103054999.0,
      "step": 49640
    },
    {
      "entropy": 5.464798641204834,
      "epoch": 4.503356313497823,
      "grad_norm": 0.96875,
      "learning_rate": 0.00031345038521670907,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21321532875299454,
      "num_tokens": 103065163.0,
      "step": 49645
    },
    {
      "entropy": 5.576402235031128,
      "epoch": 4.503809869375907,
      "grad_norm": 1.125,
      "learning_rate": 0.0003134185043620654,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22611500918865204,
      "num_tokens": 103075979.0,
      "step": 49650
    },
    {
      "entropy": 5.44485125541687,
      "epoch": 4.504263425253991,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003133866227129171,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2239288792014122,
      "num_tokens": 103086124.0,
      "step": 49655
    },
    {
      "entropy": 5.41540002822876,
      "epoch": 4.504716981132075,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031335474026992377,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21734048724174498,
      "num_tokens": 103095330.0,
      "step": 49660
    },
    {
      "entropy": 5.477388572692871,
      "epoch": 4.50517053701016,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003133228570337447,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21520823389291763,
      "num_tokens": 103106863.0,
      "step": 49665
    },
    {
      "entropy": 5.534853839874268,
      "epoch": 4.505624092888244,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003132909730050391,
      "loss": 4.986,
      "mean_token_accuracy": 0.2141174405813217,
      "num_tokens": 103117256.0,
      "step": 49670
    },
    {
      "entropy": 5.509948492050171,
      "epoch": 4.506077648766328,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00031325908818446643,
      "loss": 4.9335,
      "mean_token_accuracy": 0.2215821176767349,
      "num_tokens": 103127601.0,
      "step": 49675
    },
    {
      "entropy": 5.503597402572632,
      "epoch": 4.506531204644412,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031322720257268607,
      "loss": 4.8845,
      "mean_token_accuracy": 0.216783007979393,
      "num_tokens": 103136868.0,
      "step": 49680
    },
    {
      "entropy": 5.462332677841187,
      "epoch": 4.506984760522497,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031319531617035747,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21785227060317994,
      "num_tokens": 103146908.0,
      "step": 49685
    },
    {
      "entropy": 5.401022434234619,
      "epoch": 4.507438316400581,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031316342897813994,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21737601459026337,
      "num_tokens": 103157244.0,
      "step": 49690
    },
    {
      "entropy": 5.434913635253906,
      "epoch": 4.507891872278664,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031313154099669294,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2172268435359001,
      "num_tokens": 103168502.0,
      "step": 49695
    },
    {
      "entropy": 5.511435413360596,
      "epoch": 4.5083454281567485,
      "grad_norm": 1.0,
      "learning_rate": 0.0003130996522266761,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21635978817939758,
      "num_tokens": 103179015.0,
      "step": 49700
    },
    {
      "entropy": 5.554701709747315,
      "epoch": 4.508798984034833,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003130677626687485,
      "loss": 4.9395,
      "mean_token_accuracy": 0.22129687368869783,
      "num_tokens": 103189200.0,
      "step": 49705
    },
    {
      "entropy": 5.554745006561279,
      "epoch": 4.509252539912917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003130358723235702,
      "loss": 5.0934,
      "mean_token_accuracy": 0.20941739678382873,
      "num_tokens": 103200683.0,
      "step": 49710
    },
    {
      "entropy": 5.489160490036011,
      "epoch": 4.509706095791001,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031300398119180013,
      "loss": 4.9645,
      "mean_token_accuracy": 0.22009970992803574,
      "num_tokens": 103210992.0,
      "step": 49715
    },
    {
      "entropy": 5.457375192642212,
      "epoch": 4.510159651669086,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003129720892740981,
      "loss": 4.8694,
      "mean_token_accuracy": 0.21856247633695602,
      "num_tokens": 103220783.0,
      "step": 49720
    },
    {
      "entropy": 5.441335582733155,
      "epoch": 4.51061320754717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003129401965711236,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21973849534988404,
      "num_tokens": 103231593.0,
      "step": 49725
    },
    {
      "entropy": 5.488912296295166,
      "epoch": 4.511066763425254,
      "grad_norm": 1.0,
      "learning_rate": 0.0003129083030835361,
      "loss": 4.9638,
      "mean_token_accuracy": 0.21323495358228683,
      "num_tokens": 103242416.0,
      "step": 49730
    },
    {
      "entropy": 5.41271824836731,
      "epoch": 4.511520319303338,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031287640881199534,
      "loss": 4.8601,
      "mean_token_accuracy": 0.22667614072561265,
      "num_tokens": 103252535.0,
      "step": 49735
    },
    {
      "entropy": 5.543447256088257,
      "epoch": 4.511973875181423,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003128445137571607,
      "loss": 5.0183,
      "mean_token_accuracy": 0.20899491757154465,
      "num_tokens": 103263039.0,
      "step": 49740
    },
    {
      "entropy": 5.516839647293091,
      "epoch": 4.512427431059507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031281261791969196,
      "loss": 4.904,
      "mean_token_accuracy": 0.22166491448879241,
      "num_tokens": 103273304.0,
      "step": 49745
    },
    {
      "entropy": 5.526889133453369,
      "epoch": 4.51288098693759,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031278072130024856,
      "loss": 5.0417,
      "mean_token_accuracy": 0.21756334304809571,
      "num_tokens": 103283688.0,
      "step": 49750
    },
    {
      "entropy": 5.524950742721558,
      "epoch": 4.513334542815675,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003127488238994902,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21583550572395324,
      "num_tokens": 103293572.0,
      "step": 49755
    },
    {
      "entropy": 5.516332626342773,
      "epoch": 4.513788098693759,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003127169257180766,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21814631521701813,
      "num_tokens": 103303474.0,
      "step": 49760
    },
    {
      "entropy": 5.585920619964599,
      "epoch": 4.514241654571843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003126850267566674,
      "loss": 5.0678,
      "mean_token_accuracy": 0.20586851686239244,
      "num_tokens": 103314343.0,
      "step": 49765
    },
    {
      "entropy": 5.500597047805786,
      "epoch": 4.514695210449927,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003126531270159221,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22444769144058227,
      "num_tokens": 103323942.0,
      "step": 49770
    },
    {
      "entropy": 5.5142967224121096,
      "epoch": 4.515148766328012,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003126212264965007,
      "loss": 4.947,
      "mean_token_accuracy": 0.2164674371480942,
      "num_tokens": 103333785.0,
      "step": 49775
    },
    {
      "entropy": 5.384771251678467,
      "epoch": 4.515602322206096,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031258932519906266,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23107819259166718,
      "num_tokens": 103344184.0,
      "step": 49780
    },
    {
      "entropy": 5.505048942565918,
      "epoch": 4.51605587808418,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00031255742312426786,
      "loss": 5.0096,
      "mean_token_accuracy": 0.21746401041746138,
      "num_tokens": 103355085.0,
      "step": 49785
    },
    {
      "entropy": 5.47769021987915,
      "epoch": 4.5165094339622645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031252552027277595,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22070708274841308,
      "num_tokens": 103365942.0,
      "step": 49790
    },
    {
      "entropy": 5.4293458461761475,
      "epoch": 4.516962989840349,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003124936166452467,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2260700285434723,
      "num_tokens": 103375844.0,
      "step": 49795
    },
    {
      "entropy": 5.4751404285430905,
      "epoch": 4.517416545718433,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003124617122423399,
      "loss": 4.9829,
      "mean_token_accuracy": 0.22225179225206376,
      "num_tokens": 103385871.0,
      "step": 49800
    },
    {
      "entropy": 5.510538721084595,
      "epoch": 4.517870101596516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003124298070647154,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2161705270409584,
      "num_tokens": 103396888.0,
      "step": 49805
    },
    {
      "entropy": 5.4824260711669925,
      "epoch": 4.518323657474601,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003123979011130329,
      "loss": 4.939,
      "mean_token_accuracy": 0.2199989527463913,
      "num_tokens": 103406913.0,
      "step": 49810
    },
    {
      "entropy": 5.486189603805542,
      "epoch": 4.518777213352685,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003123659943879523,
      "loss": 4.873,
      "mean_token_accuracy": 0.22807676643133162,
      "num_tokens": 103416485.0,
      "step": 49815
    },
    {
      "entropy": 5.429030799865723,
      "epoch": 4.519230769230769,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003123340868901334,
      "loss": 4.8623,
      "mean_token_accuracy": 0.21809977144002915,
      "num_tokens": 103427476.0,
      "step": 49820
    },
    {
      "entropy": 5.431174373626709,
      "epoch": 4.519684325108853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031230217862023604,
      "loss": 4.9305,
      "mean_token_accuracy": 0.22215534001588821,
      "num_tokens": 103437500.0,
      "step": 49825
    },
    {
      "entropy": 5.455555820465088,
      "epoch": 4.520137880986938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031227026957892013,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22384431213140488,
      "num_tokens": 103447646.0,
      "step": 49830
    },
    {
      "entropy": 5.458868598937988,
      "epoch": 4.520591436865022,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031223835976684555,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2206593096256256,
      "num_tokens": 103457698.0,
      "step": 49835
    },
    {
      "entropy": 5.551089715957642,
      "epoch": 4.521044992743106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031220644918467224,
      "loss": 4.9696,
      "mean_token_accuracy": 0.22083193063735962,
      "num_tokens": 103467566.0,
      "step": 49840
    },
    {
      "entropy": 5.464757108688355,
      "epoch": 4.5214985486211905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031217453783306007,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21362858414649963,
      "num_tokens": 103477724.0,
      "step": 49845
    },
    {
      "entropy": 5.496932125091552,
      "epoch": 4.521952104499274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003121426257126689,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2167811006307602,
      "num_tokens": 103487864.0,
      "step": 49850
    },
    {
      "entropy": 5.476141834259034,
      "epoch": 4.522405660377358,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031211071282415885,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22764494121074677,
      "num_tokens": 103497229.0,
      "step": 49855
    },
    {
      "entropy": 5.431722688674927,
      "epoch": 4.522859216255442,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003120787991681898,
      "loss": 4.8496,
      "mean_token_accuracy": 0.2309449017047882,
      "num_tokens": 103506984.0,
      "step": 49860
    },
    {
      "entropy": 5.433378791809082,
      "epoch": 4.523312772133527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003120468847454217,
      "loss": 4.891,
      "mean_token_accuracy": 0.21895976215600968,
      "num_tokens": 103517287.0,
      "step": 49865
    },
    {
      "entropy": 5.441697311401367,
      "epoch": 4.523766328011611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031201496955651466,
      "loss": 4.911,
      "mean_token_accuracy": 0.21975866109132766,
      "num_tokens": 103528575.0,
      "step": 49870
    },
    {
      "entropy": 5.549579048156739,
      "epoch": 4.524219883889695,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00031198305360212854,
      "loss": 5.0354,
      "mean_token_accuracy": 0.21018802970647812,
      "num_tokens": 103538406.0,
      "step": 49875
    },
    {
      "entropy": 5.482507944107056,
      "epoch": 4.5246734397677795,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003119511368829235,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21243754029273987,
      "num_tokens": 103548738.0,
      "step": 49880
    },
    {
      "entropy": 5.43783164024353,
      "epoch": 4.525126995645864,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031191921939955947,
      "loss": 4.866,
      "mean_token_accuracy": 0.22440971583127975,
      "num_tokens": 103558755.0,
      "step": 49885
    },
    {
      "entropy": 5.5465168952941895,
      "epoch": 4.525580551523948,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003118873011526966,
      "loss": 4.946,
      "mean_token_accuracy": 0.2183450311422348,
      "num_tokens": 103569115.0,
      "step": 49890
    },
    {
      "entropy": 5.595797395706176,
      "epoch": 4.526034107402032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003118553821429949,
      "loss": 4.9962,
      "mean_token_accuracy": 0.21671897917985916,
      "num_tokens": 103580453.0,
      "step": 49895
    },
    {
      "entropy": 5.486563014984131,
      "epoch": 4.5264876632801165,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003118234623711146,
      "loss": 4.9995,
      "mean_token_accuracy": 0.21996249109506608,
      "num_tokens": 103592995.0,
      "step": 49900
    },
    {
      "entropy": 5.4331052780151365,
      "epoch": 4.5269412191582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031179154183771576,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22891646772623062,
      "num_tokens": 103603496.0,
      "step": 49905
    },
    {
      "entropy": 5.5302405834198,
      "epoch": 4.527394775036284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003117596205434583,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21903739720582963,
      "num_tokens": 103615028.0,
      "step": 49910
    },
    {
      "entropy": 5.470948076248169,
      "epoch": 4.5278483309143684,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031172769848900256,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21503289490938188,
      "num_tokens": 103625430.0,
      "step": 49915
    },
    {
      "entropy": 5.519324254989624,
      "epoch": 4.528301886792453,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003116957756750087,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2157749593257904,
      "num_tokens": 103636409.0,
      "step": 49920
    },
    {
      "entropy": 5.4788342952728275,
      "epoch": 4.528755442670537,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031166385210213684,
      "loss": 4.9917,
      "mean_token_accuracy": 0.22128874510526658,
      "num_tokens": 103646869.0,
      "step": 49925
    },
    {
      "entropy": 5.433212518692017,
      "epoch": 4.529208998548621,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031163192777104715,
      "loss": 4.7874,
      "mean_token_accuracy": 0.22686613202095032,
      "num_tokens": 103657098.0,
      "step": 49930
    },
    {
      "entropy": 5.55483603477478,
      "epoch": 4.5296625544267055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003116000026823999,
      "loss": 4.9988,
      "mean_token_accuracy": 0.2182709500193596,
      "num_tokens": 103667331.0,
      "step": 49935
    },
    {
      "entropy": 5.470140886306763,
      "epoch": 4.53011611030479,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003115680768368552,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2267426759004593,
      "num_tokens": 103676755.0,
      "step": 49940
    },
    {
      "entropy": 5.430869770050049,
      "epoch": 4.530569666182874,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003115361502350734,
      "loss": 4.865,
      "mean_token_accuracy": 0.2186673179268837,
      "num_tokens": 103686334.0,
      "step": 49945
    },
    {
      "entropy": 5.480339050292969,
      "epoch": 4.531023222060957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003115042228777148,
      "loss": 4.9895,
      "mean_token_accuracy": 0.2143478736281395,
      "num_tokens": 103697127.0,
      "step": 49950
    },
    {
      "entropy": 5.424283266067505,
      "epoch": 4.531476777939043,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031147229476543935,
      "loss": 4.9028,
      "mean_token_accuracy": 0.209726719558239,
      "num_tokens": 103707860.0,
      "step": 49955
    },
    {
      "entropy": 5.569537258148193,
      "epoch": 4.531930333817126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00031144036589890765,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20998159945011138,
      "num_tokens": 103718816.0,
      "step": 49960
    },
    {
      "entropy": 5.42869873046875,
      "epoch": 4.53238388969521,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031140843627878003,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22394371330738067,
      "num_tokens": 103729203.0,
      "step": 49965
    },
    {
      "entropy": 5.55342206954956,
      "epoch": 4.5328374455732945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031137650590571655,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21691617369651794,
      "num_tokens": 103740830.0,
      "step": 49970
    },
    {
      "entropy": 5.489131927490234,
      "epoch": 4.533291001451379,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003113445747803776,
      "loss": 4.964,
      "mean_token_accuracy": 0.2165638491511345,
      "num_tokens": 103752881.0,
      "step": 49975
    },
    {
      "entropy": 5.510800981521607,
      "epoch": 4.533744557329463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003113126429034237,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2165696829557419,
      "num_tokens": 103762875.0,
      "step": 49980
    },
    {
      "entropy": 5.506253385543824,
      "epoch": 4.534198113207547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031128071027551504,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2188371643424034,
      "num_tokens": 103772147.0,
      "step": 49985
    },
    {
      "entropy": 5.411963176727295,
      "epoch": 4.5346516690856316,
      "grad_norm": 1.140625,
      "learning_rate": 0.000311248776897312,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22018184512853622,
      "num_tokens": 103782904.0,
      "step": 49990
    },
    {
      "entropy": 5.4079796314239506,
      "epoch": 4.535105224963716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031121684276947504,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21925780773162842,
      "num_tokens": 103793202.0,
      "step": 49995
    },
    {
      "entropy": 5.623546361923218,
      "epoch": 4.5355587808418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003111849078926646,
      "loss": 5.14,
      "mean_token_accuracy": 0.20973799824714662,
      "num_tokens": 103805796.0,
      "step": 50000
    },
    {
      "entropy": 5.589092540740967,
      "epoch": 4.5360123367198835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000311152972267541,
      "loss": 4.9741,
      "mean_token_accuracy": 0.2132863625884056,
      "num_tokens": 103816622.0,
      "step": 50005
    },
    {
      "entropy": 5.439120197296143,
      "epoch": 4.536465892597968,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003111210358947647,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22865354865789414,
      "num_tokens": 103826174.0,
      "step": 50010
    },
    {
      "entropy": 5.49043025970459,
      "epoch": 4.536919448476052,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003110890987749962,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2216029718518257,
      "num_tokens": 103836217.0,
      "step": 50015
    },
    {
      "entropy": 5.460951948165894,
      "epoch": 4.537373004354136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003110571609088959,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22778182178735734,
      "num_tokens": 103846661.0,
      "step": 50020
    },
    {
      "entropy": 5.509233474731445,
      "epoch": 4.5378265602322205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003110252222971244,
      "loss": 5.0039,
      "mean_token_accuracy": 0.2132986903190613,
      "num_tokens": 103857323.0,
      "step": 50025
    },
    {
      "entropy": 5.548218059539795,
      "epoch": 4.538280116110305,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003109932829403421,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2122109279036522,
      "num_tokens": 103867409.0,
      "step": 50030
    },
    {
      "entropy": 5.547846078872681,
      "epoch": 4.538733671988389,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003109613428392095,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21995198279619216,
      "num_tokens": 103877603.0,
      "step": 50035
    },
    {
      "entropy": 5.480932331085205,
      "epoch": 4.539187227866473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031092940199438715,
      "loss": 4.9916,
      "mean_token_accuracy": 0.2159641906619072,
      "num_tokens": 103888236.0,
      "step": 50040
    },
    {
      "entropy": 5.455902433395385,
      "epoch": 4.539640783744558,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003108974604065357,
      "loss": 4.9484,
      "mean_token_accuracy": 0.21946577727794647,
      "num_tokens": 103898474.0,
      "step": 50045
    },
    {
      "entropy": 5.499585723876953,
      "epoch": 4.540094339622642,
      "grad_norm": 0.890625,
      "learning_rate": 0.00031086551807631553,
      "loss": 4.925,
      "mean_token_accuracy": 0.22433449029922486,
      "num_tokens": 103909854.0,
      "step": 50050
    },
    {
      "entropy": 5.495049047470093,
      "epoch": 4.540547895500726,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031083357500438736,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21673876494169236,
      "num_tokens": 103919227.0,
      "step": 50055
    },
    {
      "entropy": 5.441840362548828,
      "epoch": 4.5410014513788095,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031080163119141167,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2279090940952301,
      "num_tokens": 103929022.0,
      "step": 50060
    },
    {
      "entropy": 5.527106761932373,
      "epoch": 4.541455007256894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031076968663804917,
      "loss": 4.9427,
      "mean_token_accuracy": 0.21751125156879425,
      "num_tokens": 103939441.0,
      "step": 50065
    },
    {
      "entropy": 5.428796815872192,
      "epoch": 4.541908563134978,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031073774134496047,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22010182589292526,
      "num_tokens": 103950312.0,
      "step": 50070
    },
    {
      "entropy": 5.498596906661987,
      "epoch": 4.542362119013062,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003107057953128061,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21433417201042176,
      "num_tokens": 103960415.0,
      "step": 50075
    },
    {
      "entropy": 5.468930053710937,
      "epoch": 4.542815674891147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003106738485422468,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2272128716111183,
      "num_tokens": 103971183.0,
      "step": 50080
    },
    {
      "entropy": 5.584605264663696,
      "epoch": 4.543269230769231,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003106419010339433,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2128118097782135,
      "num_tokens": 103982120.0,
      "step": 50085
    },
    {
      "entropy": 5.487476539611817,
      "epoch": 4.543722786647315,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031060995278855616,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22459597140550613,
      "num_tokens": 103992116.0,
      "step": 50090
    },
    {
      "entropy": 5.42745418548584,
      "epoch": 4.544176342525399,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031057800380674604,
      "loss": 4.9561,
      "mean_token_accuracy": 0.221602226793766,
      "num_tokens": 104002646.0,
      "step": 50095
    },
    {
      "entropy": 5.509330892562867,
      "epoch": 4.544629898403484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003105460540891738,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21450336128473282,
      "num_tokens": 104013158.0,
      "step": 50100
    },
    {
      "entropy": 5.456026697158814,
      "epoch": 4.545083454281567,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003105141036365001,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21666991263628005,
      "num_tokens": 104023485.0,
      "step": 50105
    },
    {
      "entropy": 5.499708271026611,
      "epoch": 4.545537010159651,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003104821524493857,
      "loss": 4.9583,
      "mean_token_accuracy": 0.22079586386680602,
      "num_tokens": 104034588.0,
      "step": 50110
    },
    {
      "entropy": 5.518093776702881,
      "epoch": 4.5459905660377355,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003104502005284913,
      "loss": 4.936,
      "mean_token_accuracy": 0.2183978497982025,
      "num_tokens": 104045081.0,
      "step": 50115
    },
    {
      "entropy": 5.509505558013916,
      "epoch": 4.54644412191582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003104182478744778,
      "loss": 4.9077,
      "mean_token_accuracy": 0.21880649030208588,
      "num_tokens": 104056166.0,
      "step": 50120
    },
    {
      "entropy": 5.488030481338501,
      "epoch": 4.546897677793904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003103862944880058,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22726378142833709,
      "num_tokens": 104066699.0,
      "step": 50125
    },
    {
      "entropy": 5.4784629344940186,
      "epoch": 4.547351233671988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003103543403697363,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21773227602243422,
      "num_tokens": 104077061.0,
      "step": 50130
    },
    {
      "entropy": 5.50829815864563,
      "epoch": 4.547804789550073,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003103223855203301,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21395813971757888,
      "num_tokens": 104087877.0,
      "step": 50135
    },
    {
      "entropy": 5.5679856300354,
      "epoch": 4.548258345428157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031029042994044784,
      "loss": 5.0368,
      "mean_token_accuracy": 0.20855233818292618,
      "num_tokens": 104099483.0,
      "step": 50140
    },
    {
      "entropy": 5.447189521789551,
      "epoch": 4.548711901306241,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031025847363075056,
      "loss": 4.899,
      "mean_token_accuracy": 0.21648306399583817,
      "num_tokens": 104109530.0,
      "step": 50145
    },
    {
      "entropy": 5.482365131378174,
      "epoch": 4.549165457184325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031022651659189907,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21751505881547928,
      "num_tokens": 104119610.0,
      "step": 50150
    },
    {
      "entropy": 5.573596572875976,
      "epoch": 4.54961901306241,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003101945588245541,
      "loss": 4.9997,
      "mean_token_accuracy": 0.2134975880384445,
      "num_tokens": 104130337.0,
      "step": 50155
    },
    {
      "entropy": 5.587914848327637,
      "epoch": 4.550072568940493,
      "grad_norm": 1.0,
      "learning_rate": 0.0003101626003293768,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2157904848456383,
      "num_tokens": 104139462.0,
      "step": 50160
    },
    {
      "entropy": 5.433866548538208,
      "epoch": 4.550526124818577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003101306411070279,
      "loss": 4.847,
      "mean_token_accuracy": 0.232054902613163,
      "num_tokens": 104148640.0,
      "step": 50165
    },
    {
      "entropy": 5.451327991485596,
      "epoch": 4.550979680696662,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031009868115816847,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2238710790872574,
      "num_tokens": 104159400.0,
      "step": 50170
    },
    {
      "entropy": 5.485476016998291,
      "epoch": 4.551433236574746,
      "grad_norm": 1.015625,
      "learning_rate": 0.00031006672048345936,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22296279221773146,
      "num_tokens": 104170149.0,
      "step": 50175
    },
    {
      "entropy": 5.459910488128662,
      "epoch": 4.55188679245283,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003100347590835615,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22469817847013474,
      "num_tokens": 104180787.0,
      "step": 50180
    },
    {
      "entropy": 5.5494420528411865,
      "epoch": 4.552340348330914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003100027969591359,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21814551949501038,
      "num_tokens": 104190961.0,
      "step": 50185
    },
    {
      "entropy": 5.466485166549683,
      "epoch": 4.552793904208999,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030997083411084356,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2232416346669197,
      "num_tokens": 104201886.0,
      "step": 50190
    },
    {
      "entropy": 5.523044109344482,
      "epoch": 4.553247460087083,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003099388705393454,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2269026443362236,
      "num_tokens": 104212834.0,
      "step": 50195
    },
    {
      "entropy": 5.479256677627563,
      "epoch": 4.553701015965167,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003099069062453025,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22299642115831375,
      "num_tokens": 104222384.0,
      "step": 50200
    },
    {
      "entropy": 5.429510450363159,
      "epoch": 4.5541545718432515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030987494122937604,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22288387417793273,
      "num_tokens": 104232971.0,
      "step": 50205
    },
    {
      "entropy": 5.501441526412964,
      "epoch": 4.554608127721336,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003098429754922267,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22006368935108184,
      "num_tokens": 104245239.0,
      "step": 50210
    },
    {
      "entropy": 5.526629209518433,
      "epoch": 4.555061683599419,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030981100903451587,
      "loss": 4.9769,
      "mean_token_accuracy": 0.215510031580925,
      "num_tokens": 104256323.0,
      "step": 50215
    },
    {
      "entropy": 5.525943946838379,
      "epoch": 4.555515239477503,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003097790418569045,
      "loss": 4.967,
      "mean_token_accuracy": 0.21754927784204484,
      "num_tokens": 104267106.0,
      "step": 50220
    },
    {
      "entropy": 5.48503303527832,
      "epoch": 4.555968795355588,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003097470739600536,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22279607355594636,
      "num_tokens": 104277336.0,
      "step": 50225
    },
    {
      "entropy": 5.479040718078613,
      "epoch": 4.556422351233672,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003097151053446244,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22556447386741638,
      "num_tokens": 104288809.0,
      "step": 50230
    },
    {
      "entropy": 5.4847083568573,
      "epoch": 4.556875907111756,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003096831360112781,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2126552864909172,
      "num_tokens": 104299434.0,
      "step": 50235
    },
    {
      "entropy": 5.45778112411499,
      "epoch": 4.55732946298984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003096511659606756,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2208981230854988,
      "num_tokens": 104309426.0,
      "step": 50240
    },
    {
      "entropy": 5.443497610092163,
      "epoch": 4.557783018867925,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003096191951934781,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22090920358896254,
      "num_tokens": 104319530.0,
      "step": 50245
    },
    {
      "entropy": 5.456666326522827,
      "epoch": 4.558236574746009,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003095872237103468,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22213948220014573,
      "num_tokens": 104329198.0,
      "step": 50250
    },
    {
      "entropy": 5.547898006439209,
      "epoch": 4.558690130624093,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030955525151194304,
      "loss": 5.0409,
      "mean_token_accuracy": 0.21508552581071855,
      "num_tokens": 104340481.0,
      "step": 50255
    },
    {
      "entropy": 5.518746948242187,
      "epoch": 4.559143686502177,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003095232785989277,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22366082519292832,
      "num_tokens": 104350382.0,
      "step": 50260
    },
    {
      "entropy": 5.461147260665894,
      "epoch": 4.559597242380261,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003094913049719623,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22269917875528336,
      "num_tokens": 104360967.0,
      "step": 50265
    },
    {
      "entropy": 5.478045225143433,
      "epoch": 4.560050798258345,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030945933063170793,
      "loss": 4.9253,
      "mean_token_accuracy": 0.22396702766418458,
      "num_tokens": 104371283.0,
      "step": 50270
    },
    {
      "entropy": 5.504816484451294,
      "epoch": 4.560504354136429,
      "grad_norm": 0.953125,
      "learning_rate": 0.00030942735557882575,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21921756863594055,
      "num_tokens": 104381545.0,
      "step": 50275
    },
    {
      "entropy": 5.577618980407715,
      "epoch": 4.560957910014514,
      "grad_norm": 1.125,
      "learning_rate": 0.00030939537981397716,
      "loss": 5.0171,
      "mean_token_accuracy": 0.2105001136660576,
      "num_tokens": 104392533.0,
      "step": 50280
    },
    {
      "entropy": 5.511122369766236,
      "epoch": 4.561411465892598,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030936340333782326,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21634143441915513,
      "num_tokens": 104402230.0,
      "step": 50285
    },
    {
      "entropy": 5.507300090789795,
      "epoch": 4.561865021770682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003093314261510254,
      "loss": 4.9259,
      "mean_token_accuracy": 0.22001306265592574,
      "num_tokens": 104412029.0,
      "step": 50290
    },
    {
      "entropy": 5.451783752441406,
      "epoch": 4.5623185776487665,
      "grad_norm": 1.0625,
      "learning_rate": 0.000309299448254245,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21852934658527373,
      "num_tokens": 104422107.0,
      "step": 50295
    },
    {
      "entropy": 5.505238246917725,
      "epoch": 4.562772133526851,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030926746964814324,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2185912847518921,
      "num_tokens": 104432073.0,
      "step": 50300
    },
    {
      "entropy": 5.460331153869629,
      "epoch": 4.563225689404935,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030923549033338147,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22406792342662812,
      "num_tokens": 104441998.0,
      "step": 50305
    },
    {
      "entropy": 5.368362998962402,
      "epoch": 4.563679245283019,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000309203510310621,
      "loss": 4.796,
      "mean_token_accuracy": 0.23127072900533677,
      "num_tokens": 104451474.0,
      "step": 50310
    },
    {
      "entropy": 5.43036584854126,
      "epoch": 4.564132801161103,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003091715295805233,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2268666371703148,
      "num_tokens": 104461351.0,
      "step": 50315
    },
    {
      "entropy": 5.524000644683838,
      "epoch": 4.564586357039187,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003091395481437496,
      "loss": 4.912,
      "mean_token_accuracy": 0.22272174209356307,
      "num_tokens": 104471013.0,
      "step": 50320
    },
    {
      "entropy": 5.526110029220581,
      "epoch": 4.565039912917271,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003091075660009614,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2181393548846245,
      "num_tokens": 104481863.0,
      "step": 50325
    },
    {
      "entropy": 5.489814186096192,
      "epoch": 4.5654934687953554,
      "grad_norm": 0.953125,
      "learning_rate": 0.00030907558315282,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21388644129037857,
      "num_tokens": 104492718.0,
      "step": 50330
    },
    {
      "entropy": 5.478087663650513,
      "epoch": 4.56594702467344,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030904359959998685,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2187313973903656,
      "num_tokens": 104502837.0,
      "step": 50335
    },
    {
      "entropy": 5.5470277786254885,
      "epoch": 4.566400580551524,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003090116153431234,
      "loss": 4.9885,
      "mean_token_accuracy": 0.22344374656677246,
      "num_tokens": 104514202.0,
      "step": 50340
    },
    {
      "entropy": 5.532456588745117,
      "epoch": 4.566854136429608,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000308979630382891,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2133808836340904,
      "num_tokens": 104525037.0,
      "step": 50345
    },
    {
      "entropy": 5.415830278396607,
      "epoch": 4.5673076923076925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030894764471995137,
      "loss": 4.8747,
      "mean_token_accuracy": 0.21751632541418076,
      "num_tokens": 104534883.0,
      "step": 50350
    },
    {
      "entropy": 5.444731569290161,
      "epoch": 4.567761248185777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003089156583549657,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22105929255485535,
      "num_tokens": 104544845.0,
      "step": 50355
    },
    {
      "entropy": 5.507034730911255,
      "epoch": 4.568214804063861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003088836712885955,
      "loss": 4.9877,
      "mean_token_accuracy": 0.2145534411072731,
      "num_tokens": 104555654.0,
      "step": 50360
    },
    {
      "entropy": 5.424383926391601,
      "epoch": 4.568668359941945,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030885168352150243,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21222768276929854,
      "num_tokens": 104565015.0,
      "step": 50365
    },
    {
      "entropy": 5.406116724014282,
      "epoch": 4.569121915820029,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030881969505434795,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2218663901090622,
      "num_tokens": 104575934.0,
      "step": 50370
    },
    {
      "entropy": 5.4311905860900875,
      "epoch": 4.569575471698113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030878770588779335,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22559343427419662,
      "num_tokens": 104585915.0,
      "step": 50375
    },
    {
      "entropy": 5.537026643753052,
      "epoch": 4.570029027576197,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030875571602250064,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21583542674779893,
      "num_tokens": 104595605.0,
      "step": 50380
    },
    {
      "entropy": 5.52308201789856,
      "epoch": 4.5704825834542815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000308723725459131,
      "loss": 5.0471,
      "mean_token_accuracy": 0.2032479837536812,
      "num_tokens": 104608251.0,
      "step": 50385
    },
    {
      "entropy": 5.529435920715332,
      "epoch": 4.570936139332366,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003086917341983461,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22705529481172562,
      "num_tokens": 104618586.0,
      "step": 50390
    },
    {
      "entropy": 5.452147579193115,
      "epoch": 4.57138969521045,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003086597422408076,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21952458173036576,
      "num_tokens": 104629718.0,
      "step": 50395
    },
    {
      "entropy": 5.455179214477539,
      "epoch": 4.571843251088534,
      "grad_norm": 1.015625,
      "learning_rate": 0.000308627749587177,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22273576855659485,
      "num_tokens": 104639923.0,
      "step": 50400
    },
    {
      "entropy": 5.480712842941284,
      "epoch": 4.5722968069666186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000308595756238116,
      "loss": 4.9561,
      "mean_token_accuracy": 0.21706577241420746,
      "num_tokens": 104650890.0,
      "step": 50405
    },
    {
      "entropy": 5.4906532764434814,
      "epoch": 4.572750362844703,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003085637621942862,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2254445418715477,
      "num_tokens": 104661231.0,
      "step": 50410
    },
    {
      "entropy": 5.467855501174927,
      "epoch": 4.573203918722786,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00030853176745634925,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2276921883225441,
      "num_tokens": 104671641.0,
      "step": 50415
    },
    {
      "entropy": 5.498576641082764,
      "epoch": 4.5736574746008705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003084997720249668,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21981891393661498,
      "num_tokens": 104682310.0,
      "step": 50420
    },
    {
      "entropy": 5.516867160797119,
      "epoch": 4.574111030478955,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030846777590080057,
      "loss": 5.0036,
      "mean_token_accuracy": 0.2168795645236969,
      "num_tokens": 104692414.0,
      "step": 50425
    },
    {
      "entropy": 5.464972305297851,
      "epoch": 4.574564586357039,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030843577908451215,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21163810044527054,
      "num_tokens": 104703617.0,
      "step": 50430
    },
    {
      "entropy": 5.407656240463257,
      "epoch": 4.575018142235123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003084037815767633,
      "loss": 4.9005,
      "mean_token_accuracy": 0.22251663208007813,
      "num_tokens": 104713494.0,
      "step": 50435
    },
    {
      "entropy": 5.469925212860107,
      "epoch": 4.5754716981132075,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003083717833782158,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22073530554771423,
      "num_tokens": 104723349.0,
      "step": 50440
    },
    {
      "entropy": 5.455719661712647,
      "epoch": 4.575925253991292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003083397844895312,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23091717213392257,
      "num_tokens": 104732903.0,
      "step": 50445
    },
    {
      "entropy": 5.421713542938233,
      "epoch": 4.576378809869376,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003083077849113715,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22142132073640824,
      "num_tokens": 104742533.0,
      "step": 50450
    },
    {
      "entropy": 5.430556344985962,
      "epoch": 4.57683236574746,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030827578464439823,
      "loss": 4.8169,
      "mean_token_accuracy": 0.21743826419115067,
      "num_tokens": 104752161.0,
      "step": 50455
    },
    {
      "entropy": 5.427007818222046,
      "epoch": 4.577285921625545,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030824378368927324,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22843188792467117,
      "num_tokens": 104762630.0,
      "step": 50460
    },
    {
      "entropy": 5.465293645858765,
      "epoch": 4.577739477503629,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030821178204665836,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22370698153972626,
      "num_tokens": 104773529.0,
      "step": 50465
    },
    {
      "entropy": 5.415405178070069,
      "epoch": 4.578193033381712,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003081797797172154,
      "loss": 4.8525,
      "mean_token_accuracy": 0.21742700785398483,
      "num_tokens": 104783386.0,
      "step": 50470
    },
    {
      "entropy": 5.480838489532471,
      "epoch": 4.5786465892597965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003081477767016061,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21698750406503678,
      "num_tokens": 104793254.0,
      "step": 50475
    },
    {
      "entropy": 5.499848937988281,
      "epoch": 4.579100145137881,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030811577300049226,
      "loss": 4.9482,
      "mean_token_accuracy": 0.22167288810014724,
      "num_tokens": 104804345.0,
      "step": 50480
    },
    {
      "entropy": 5.473322439193725,
      "epoch": 4.579553701015965,
      "grad_norm": 1.0,
      "learning_rate": 0.0003080837686145359,
      "loss": 4.8934,
      "mean_token_accuracy": 0.2202608108520508,
      "num_tokens": 104814891.0,
      "step": 50485
    },
    {
      "entropy": 5.46420431137085,
      "epoch": 4.580007256894049,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003080517635443987,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21848602294921876,
      "num_tokens": 104825015.0,
      "step": 50490
    },
    {
      "entropy": 5.531939315795898,
      "epoch": 4.580460812772134,
      "grad_norm": 1.0,
      "learning_rate": 0.0003080197577907427,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21173185706138611,
      "num_tokens": 104835735.0,
      "step": 50495
    },
    {
      "entropy": 5.476615381240845,
      "epoch": 4.580914368650218,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030798775135422966,
      "loss": 4.946,
      "mean_token_accuracy": 0.21565124839544297,
      "num_tokens": 104847484.0,
      "step": 50500
    },
    {
      "entropy": 5.465299320220947,
      "epoch": 4.581367924528302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003079557442355215,
      "loss": 4.9502,
      "mean_token_accuracy": 0.22009094655513764,
      "num_tokens": 104857783.0,
      "step": 50505
    },
    {
      "entropy": 5.427747583389282,
      "epoch": 4.581821480406386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003079237364352801,
      "loss": 4.8608,
      "mean_token_accuracy": 0.23109417408704758,
      "num_tokens": 104867498.0,
      "step": 50510
    },
    {
      "entropy": 5.491675949096679,
      "epoch": 4.58227503628447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003078917279541675,
      "loss": 4.932,
      "mean_token_accuracy": 0.22292470186948776,
      "num_tokens": 104876505.0,
      "step": 50515
    },
    {
      "entropy": 5.524939775466919,
      "epoch": 4.582728592162555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003078597187928456,
      "loss": 4.952,
      "mean_token_accuracy": 0.21169041693210602,
      "num_tokens": 104886932.0,
      "step": 50520
    },
    {
      "entropy": 5.509853839874268,
      "epoch": 4.583182148040638,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003078277089519764,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22423761636018752,
      "num_tokens": 104896733.0,
      "step": 50525
    },
    {
      "entropy": 5.509203863143921,
      "epoch": 4.5836357039187225,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030779569843222165,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20996283143758773,
      "num_tokens": 104906846.0,
      "step": 50530
    },
    {
      "entropy": 5.383128118515015,
      "epoch": 4.584089259796807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030776368723424365,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23835154473781586,
      "num_tokens": 104916136.0,
      "step": 50535
    },
    {
      "entropy": 5.463440418243408,
      "epoch": 4.584542815674891,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003077316753587042,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22272311747074128,
      "num_tokens": 104926145.0,
      "step": 50540
    },
    {
      "entropy": 5.463179588317871,
      "epoch": 4.584996371552975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030769966280626526,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20894578397274016,
      "num_tokens": 104935576.0,
      "step": 50545
    },
    {
      "entropy": 5.433211803436279,
      "epoch": 4.58544992743106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030766764957758906,
      "loss": 4.809,
      "mean_token_accuracy": 0.2284143626689911,
      "num_tokens": 104944846.0,
      "step": 50550
    },
    {
      "entropy": 5.456935882568359,
      "epoch": 4.585903483309144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030763563567333754,
      "loss": 4.8619,
      "mean_token_accuracy": 0.22537506222724915,
      "num_tokens": 104954604.0,
      "step": 50555
    },
    {
      "entropy": 5.470808696746826,
      "epoch": 4.586357039187228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030760362109417277,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22082064151763917,
      "num_tokens": 104964858.0,
      "step": 50560
    },
    {
      "entropy": 5.535459280014038,
      "epoch": 4.586810595065312,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003075716058407568,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21519654840230942,
      "num_tokens": 104974997.0,
      "step": 50565
    },
    {
      "entropy": 5.466248559951782,
      "epoch": 4.587264150943396,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030753958991375176,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22813080996274948,
      "num_tokens": 104984585.0,
      "step": 50570
    },
    {
      "entropy": 5.459557771682739,
      "epoch": 4.58771770682148,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030750757331381964,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2235298365354538,
      "num_tokens": 104994843.0,
      "step": 50575
    },
    {
      "entropy": 5.496619939804077,
      "epoch": 4.588171262699564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003074755560416226,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21718664914369584,
      "num_tokens": 105006082.0,
      "step": 50580
    },
    {
      "entropy": 5.464487075805664,
      "epoch": 4.588624818577649,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030744353809782283,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22321270108222963,
      "num_tokens": 105016704.0,
      "step": 50585
    },
    {
      "entropy": 5.535545349121094,
      "epoch": 4.589078374455733,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030741151948308237,
      "loss": 5.0091,
      "mean_token_accuracy": 0.2170875519514084,
      "num_tokens": 105027810.0,
      "step": 50590
    },
    {
      "entropy": 5.440300035476684,
      "epoch": 4.589531930333817,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003073795001980635,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22664288878440858,
      "num_tokens": 105037488.0,
      "step": 50595
    },
    {
      "entropy": 5.500630950927734,
      "epoch": 4.589985486211901,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003073474802434282,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21915352046489717,
      "num_tokens": 105047461.0,
      "step": 50600
    },
    {
      "entropy": 5.542118263244629,
      "epoch": 4.590439042089986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000307315459619839,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21990981847047805,
      "num_tokens": 105058020.0,
      "step": 50605
    },
    {
      "entropy": 5.583900356292725,
      "epoch": 4.59089259796807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003072834383279576,
      "loss": 4.9719,
      "mean_token_accuracy": 0.2142475202679634,
      "num_tokens": 105069638.0,
      "step": 50610
    },
    {
      "entropy": 5.480777788162231,
      "epoch": 4.591346153846154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003072514163684466,
      "loss": 4.9282,
      "mean_token_accuracy": 0.22239864021539688,
      "num_tokens": 105078686.0,
      "step": 50615
    },
    {
      "entropy": 5.4721057415008545,
      "epoch": 4.5917997097242385,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003072193937419681,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21990518569946288,
      "num_tokens": 105089739.0,
      "step": 50620
    },
    {
      "entropy": 5.494131278991699,
      "epoch": 4.592253265602322,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030718737044918435,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2216997653245926,
      "num_tokens": 105100673.0,
      "step": 50625
    },
    {
      "entropy": 5.462527942657471,
      "epoch": 4.592706821480406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030715534649075745,
      "loss": 4.9149,
      "mean_token_accuracy": 0.22281045019626616,
      "num_tokens": 105110750.0,
      "step": 50630
    },
    {
      "entropy": 5.391985559463501,
      "epoch": 4.59316037735849,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030712332186734985,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23032145351171493,
      "num_tokens": 105120809.0,
      "step": 50635
    },
    {
      "entropy": 5.504466867446899,
      "epoch": 4.593613933236575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003070912965796238,
      "loss": 4.937,
      "mean_token_accuracy": 0.21921930313110352,
      "num_tokens": 105131106.0,
      "step": 50640
    },
    {
      "entropy": 5.52215347290039,
      "epoch": 4.594067489114659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030705927062824154,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21810335516929627,
      "num_tokens": 105140970.0,
      "step": 50645
    },
    {
      "entropy": 5.5055357933044435,
      "epoch": 4.594521044992743,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003070272440138654,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2113129660487175,
      "num_tokens": 105152429.0,
      "step": 50650
    },
    {
      "entropy": 5.543699598312378,
      "epoch": 4.594974600870827,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003069952167371577,
      "loss": 4.9986,
      "mean_token_accuracy": 0.21372606307268144,
      "num_tokens": 105163534.0,
      "step": 50655
    },
    {
      "entropy": 5.485090351104736,
      "epoch": 4.595428156748912,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030696318879878064,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22359406799077988,
      "num_tokens": 105172798.0,
      "step": 50660
    },
    {
      "entropy": 5.535303926467895,
      "epoch": 4.595881712626996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030693116019939686,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2196369782090187,
      "num_tokens": 105182576.0,
      "step": 50665
    },
    {
      "entropy": 5.452533102035522,
      "epoch": 4.596335268505079,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030689913093966847,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21989029198884963,
      "num_tokens": 105192070.0,
      "step": 50670
    },
    {
      "entropy": 5.474808216094971,
      "epoch": 4.5967888243831645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000306867101020258,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2236279919743538,
      "num_tokens": 105203151.0,
      "step": 50675
    },
    {
      "entropy": 5.521759939193726,
      "epoch": 4.597242380261248,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003068350704418277,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21444423049688338,
      "num_tokens": 105213217.0,
      "step": 50680
    },
    {
      "entropy": 5.436547660827637,
      "epoch": 4.597695936139332,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030680303920504006,
      "loss": 4.825,
      "mean_token_accuracy": 0.22654947489500046,
      "num_tokens": 105223450.0,
      "step": 50685
    },
    {
      "entropy": 5.523337554931641,
      "epoch": 4.598149492017416,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003067710073105574,
      "loss": 5.0158,
      "mean_token_accuracy": 0.21612521708011628,
      "num_tokens": 105234705.0,
      "step": 50690
    },
    {
      "entropy": 5.604640150070191,
      "epoch": 4.598603047895501,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030673897475904223,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1962705969810486,
      "num_tokens": 105245042.0,
      "step": 50695
    },
    {
      "entropy": 5.542986059188843,
      "epoch": 4.599056603773585,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00030670694155115704,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2160591870546341,
      "num_tokens": 105255830.0,
      "step": 50700
    },
    {
      "entropy": 5.452032852172851,
      "epoch": 4.599510159651669,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030667490768756416,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21843077540397643,
      "num_tokens": 105265844.0,
      "step": 50705
    },
    {
      "entropy": 5.462486600875854,
      "epoch": 4.5999637155297535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003066428731689262,
      "loss": 4.944,
      "mean_token_accuracy": 0.21992552578449248,
      "num_tokens": 105276846.0,
      "step": 50710
    },
    {
      "entropy": 5.544265651702881,
      "epoch": 4.600417271407838,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030661083799590544,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21719330549240112,
      "num_tokens": 105286740.0,
      "step": 50715
    },
    {
      "entropy": 5.445932149887085,
      "epoch": 4.600870827285922,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030657880216916454,
      "loss": 4.9183,
      "mean_token_accuracy": 0.22138866186141967,
      "num_tokens": 105296710.0,
      "step": 50720
    },
    {
      "entropy": 5.491915321350097,
      "epoch": 4.601324383164005,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000306546765689366,
      "loss": 4.9553,
      "mean_token_accuracy": 0.22368945628404618,
      "num_tokens": 105307076.0,
      "step": 50725
    },
    {
      "entropy": 5.508518218994141,
      "epoch": 4.60177793904209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003065147285571722,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22496071308851243,
      "num_tokens": 105317486.0,
      "step": 50730
    },
    {
      "entropy": 5.462547540664673,
      "epoch": 4.602231494920174,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003064826907732459,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22107092738151551,
      "num_tokens": 105327771.0,
      "step": 50735
    },
    {
      "entropy": 5.494876718521118,
      "epoch": 4.602685050798258,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030645065233824955,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21802877336740495,
      "num_tokens": 105339191.0,
      "step": 50740
    },
    {
      "entropy": 5.555566215515137,
      "epoch": 4.6031386066763424,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030641861325284557,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21686145961284636,
      "num_tokens": 105349949.0,
      "step": 50745
    },
    {
      "entropy": 5.476305723190308,
      "epoch": 4.603592162554427,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00030638657351769673,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2129506841301918,
      "num_tokens": 105360357.0,
      "step": 50750
    },
    {
      "entropy": 5.578124332427978,
      "epoch": 4.604045718432511,
      "grad_norm": 0.984375,
      "learning_rate": 0.00030635453313346554,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2191293567419052,
      "num_tokens": 105370765.0,
      "step": 50755
    },
    {
      "entropy": 5.5294536590576175,
      "epoch": 4.604499274310595,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003063224921008146,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2172293096780777,
      "num_tokens": 105380797.0,
      "step": 50760
    },
    {
      "entropy": 5.458508014678955,
      "epoch": 4.6049528301886795,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003062904504204066,
      "loss": 4.9252,
      "mean_token_accuracy": 0.2199482500553131,
      "num_tokens": 105390653.0,
      "step": 50765
    },
    {
      "entropy": 5.5311071395874025,
      "epoch": 4.605406386066764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030625840809290404,
      "loss": 5.0799,
      "mean_token_accuracy": 0.21139737367630004,
      "num_tokens": 105401308.0,
      "step": 50770
    },
    {
      "entropy": 5.4944781303405765,
      "epoch": 4.605859941944848,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030622636511896977,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21478833109140397,
      "num_tokens": 105412412.0,
      "step": 50775
    },
    {
      "entropy": 5.611914253234863,
      "epoch": 4.606313497822931,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003061943214992662,
      "loss": 5.0177,
      "mean_token_accuracy": 0.21348086893558502,
      "num_tokens": 105424789.0,
      "step": 50780
    },
    {
      "entropy": 5.5020373344421385,
      "epoch": 4.606767053701016,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003061622772344561,
      "loss": 5.0251,
      "mean_token_accuracy": 0.22254009991884233,
      "num_tokens": 105434380.0,
      "step": 50785
    },
    {
      "entropy": 5.499947214126587,
      "epoch": 4.6072206095791,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003061302323252022,
      "loss": 4.9026,
      "mean_token_accuracy": 0.222102652490139,
      "num_tokens": 105443889.0,
      "step": 50790
    },
    {
      "entropy": 5.5457230567932125,
      "epoch": 4.607674165457184,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003060981867721671,
      "loss": 4.9078,
      "mean_token_accuracy": 0.22347194999456405,
      "num_tokens": 105454336.0,
      "step": 50795
    },
    {
      "entropy": 5.553639078140259,
      "epoch": 4.6081277213352685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030606614057601374,
      "loss": 5.0423,
      "mean_token_accuracy": 0.21030569970607757,
      "num_tokens": 105464601.0,
      "step": 50800
    },
    {
      "entropy": 5.550146818161011,
      "epoch": 4.608581277213353,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030603409373740456,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20886824280023575,
      "num_tokens": 105475265.0,
      "step": 50805
    },
    {
      "entropy": 5.443309879302978,
      "epoch": 4.609034833091437,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030600204625700234,
      "loss": 4.8385,
      "mean_token_accuracy": 0.2274029791355133,
      "num_tokens": 105485491.0,
      "step": 50810
    },
    {
      "entropy": 5.450588893890381,
      "epoch": 4.609488388969521,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030596999813547007,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21860937178134918,
      "num_tokens": 105495859.0,
      "step": 50815
    },
    {
      "entropy": 5.382685947418213,
      "epoch": 4.6099419448476056,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003059379493734702,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2243236243724823,
      "num_tokens": 105506385.0,
      "step": 50820
    },
    {
      "entropy": 5.454486417770386,
      "epoch": 4.610395500725689,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030590589997166575,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22718794345855714,
      "num_tokens": 105515829.0,
      "step": 50825
    },
    {
      "entropy": 5.468488597869873,
      "epoch": 4.610849056603773,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003058738499307193,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23135290443897247,
      "num_tokens": 105525950.0,
      "step": 50830
    },
    {
      "entropy": 5.4567045211792,
      "epoch": 4.6113026124818575,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030584179925129384,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21699070036411286,
      "num_tokens": 105536536.0,
      "step": 50835
    },
    {
      "entropy": 5.417964124679566,
      "epoch": 4.611756168359942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030580974793405213,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22965867817401886,
      "num_tokens": 105546069.0,
      "step": 50840
    },
    {
      "entropy": 5.455004692077637,
      "epoch": 4.612209724238026,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030577769597965696,
      "loss": 4.9265,
      "mean_token_accuracy": 0.2203232690691948,
      "num_tokens": 105556522.0,
      "step": 50845
    },
    {
      "entropy": 5.450741100311279,
      "epoch": 4.61266328011611,
      "grad_norm": 1.0,
      "learning_rate": 0.0003057456433887712,
      "loss": 4.8389,
      "mean_token_accuracy": 0.22245032340288162,
      "num_tokens": 105566431.0,
      "step": 50850
    },
    {
      "entropy": 5.433681964874268,
      "epoch": 4.6131168359941945,
      "grad_norm": 1.0,
      "learning_rate": 0.00030571359016205767,
      "loss": 4.8729,
      "mean_token_accuracy": 0.23037443310022354,
      "num_tokens": 105576270.0,
      "step": 50855
    },
    {
      "entropy": 5.444457578659057,
      "epoch": 4.613570391872279,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003056815363001793,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2179284080862999,
      "num_tokens": 105587144.0,
      "step": 50860
    },
    {
      "entropy": 5.441782379150391,
      "epoch": 4.614023947750363,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030564948180379897,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22214077413082123,
      "num_tokens": 105596636.0,
      "step": 50865
    },
    {
      "entropy": 5.5098060131072994,
      "epoch": 4.614477503628447,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003056174266735795,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22285690307617187,
      "num_tokens": 105605825.0,
      "step": 50870
    },
    {
      "entropy": 5.541480636596679,
      "epoch": 4.614931059506532,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003055853709101839,
      "loss": 5.0148,
      "mean_token_accuracy": 0.21604515761137008,
      "num_tokens": 105615990.0,
      "step": 50875
    },
    {
      "entropy": 5.476885890960693,
      "epoch": 4.615384615384615,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030555331451427494,
      "loss": 4.955,
      "mean_token_accuracy": 0.21428394317626953,
      "num_tokens": 105626673.0,
      "step": 50880
    },
    {
      "entropy": 5.532274341583252,
      "epoch": 4.615838171262699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030552125748651573,
      "loss": 4.9974,
      "mean_token_accuracy": 0.21496880054473877,
      "num_tokens": 105636752.0,
      "step": 50885
    },
    {
      "entropy": 5.563161134719849,
      "epoch": 4.6162917271407835,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030548919982756914,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21756286770105362,
      "num_tokens": 105647476.0,
      "step": 50890
    },
    {
      "entropy": 5.523286581039429,
      "epoch": 4.616745283018868,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00030545714153809805,
      "loss": 4.9324,
      "mean_token_accuracy": 0.22625567018985748,
      "num_tokens": 105658135.0,
      "step": 50895
    },
    {
      "entropy": 5.422634696960449,
      "epoch": 4.617198838896952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003054250826187655,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22512157559394835,
      "num_tokens": 105667802.0,
      "step": 50900
    },
    {
      "entropy": 5.545979928970337,
      "epoch": 4.617652394775036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003053930230702346,
      "loss": 4.9929,
      "mean_token_accuracy": 0.217341548204422,
      "num_tokens": 105679161.0,
      "step": 50905
    },
    {
      "entropy": 5.527423334121704,
      "epoch": 4.618105950653121,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003053609628931682,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2077125698328018,
      "num_tokens": 105688093.0,
      "step": 50910
    },
    {
      "entropy": 5.510435485839844,
      "epoch": 4.618559506531205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003053289020882293,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21426230669021606,
      "num_tokens": 105697982.0,
      "step": 50915
    },
    {
      "entropy": 5.510800123214722,
      "epoch": 4.619013062409289,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030529684065608103,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22371442914009093,
      "num_tokens": 105708729.0,
      "step": 50920
    },
    {
      "entropy": 5.425027561187744,
      "epoch": 4.619466618287373,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030526477859738635,
      "loss": 4.8658,
      "mean_token_accuracy": 0.22381467074155809,
      "num_tokens": 105718919.0,
      "step": 50925
    },
    {
      "entropy": 5.433631563186646,
      "epoch": 4.619920174165458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003052327159128083,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22696438431739807,
      "num_tokens": 105727787.0,
      "step": 50930
    },
    {
      "entropy": 5.468393087387085,
      "epoch": 4.620373730043541,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030520065260301003,
      "loss": 4.8755,
      "mean_token_accuracy": 0.21718854159116746,
      "num_tokens": 105738282.0,
      "step": 50935
    },
    {
      "entropy": 5.496393823623658,
      "epoch": 4.620827285921625,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003051685886686546,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21910899728536606,
      "num_tokens": 105748786.0,
      "step": 50940
    },
    {
      "entropy": 5.479892206192017,
      "epoch": 4.6212808417997095,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000305136524110405,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21780939102172853,
      "num_tokens": 105759156.0,
      "step": 50945
    },
    {
      "entropy": 5.467916202545166,
      "epoch": 4.621734397677794,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003051044589289245,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2130289316177368,
      "num_tokens": 105769434.0,
      "step": 50950
    },
    {
      "entropy": 5.570690536499024,
      "epoch": 4.622187953555878,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030507239312487595,
      "loss": 5.0715,
      "mean_token_accuracy": 0.20926281809806824,
      "num_tokens": 105779408.0,
      "step": 50955
    },
    {
      "entropy": 5.587053680419922,
      "epoch": 4.622641509433962,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030504032669892276,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21613000333309174,
      "num_tokens": 105789517.0,
      "step": 50960
    },
    {
      "entropy": 5.517906236648559,
      "epoch": 4.623095065312047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030500825965172797,
      "loss": 4.9282,
      "mean_token_accuracy": 0.22213821709156037,
      "num_tokens": 105799823.0,
      "step": 50965
    },
    {
      "entropy": 5.479869174957275,
      "epoch": 4.623548621190131,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030497619198395466,
      "loss": 4.9399,
      "mean_token_accuracy": 0.22245082557201384,
      "num_tokens": 105810038.0,
      "step": 50970
    },
    {
      "entropy": 5.515698099136353,
      "epoch": 4.624002177068215,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030494412369626615,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2184402495622635,
      "num_tokens": 105819940.0,
      "step": 50975
    },
    {
      "entropy": 5.485290145874023,
      "epoch": 4.6244557329462985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003049120547893255,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2240385890007019,
      "num_tokens": 105829829.0,
      "step": 50980
    },
    {
      "entropy": 5.452076435089111,
      "epoch": 4.624909288824383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003048799852637958,
      "loss": 4.8844,
      "mean_token_accuracy": 0.227901329100132,
      "num_tokens": 105839463.0,
      "step": 50985
    },
    {
      "entropy": 5.578142213821411,
      "epoch": 4.625362844702467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030484791512034063,
      "loss": 5.0964,
      "mean_token_accuracy": 0.20525910556316376,
      "num_tokens": 105849916.0,
      "step": 50990
    },
    {
      "entropy": 5.500504350662231,
      "epoch": 4.625816400580551,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003048158443596228,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21924666464328765,
      "num_tokens": 105860699.0,
      "step": 50995
    },
    {
      "entropy": 5.516533470153808,
      "epoch": 4.626269956458636,
      "grad_norm": 1.125,
      "learning_rate": 0.0003047837729823058,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21723813861608504,
      "num_tokens": 105870332.0,
      "step": 51000
    },
    {
      "epoch": 4.626269956458636,
      "eval_entropy": 5.1879581874312315,
      "eval_loss": 5.027714252471924,
      "eval_mean_token_accuracy": 0.2230025445831712,
      "eval_num_tokens": 105870332.0,
      "eval_runtime": 26.6716,
      "eval_samples_per_second": 1325.752,
      "eval_steps_per_second": 165.719,
      "step": 51000
    },
    {
      "entropy": 5.421049404144287,
      "epoch": 4.62672351233672,
      "grad_norm": 1.125,
      "learning_rate": 0.0003047517009890527,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22683272361755372,
      "num_tokens": 105879364.0,
      "step": 51005
    },
    {
      "entropy": 5.512977886199951,
      "epoch": 4.627177068214804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030471962838052693,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2118483915925026,
      "num_tokens": 105889554.0,
      "step": 51010
    },
    {
      "entropy": 5.476587724685669,
      "epoch": 4.627630624092888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003046875551573916,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2309255912899971,
      "num_tokens": 105900426.0,
      "step": 51015
    },
    {
      "entropy": 5.334587860107422,
      "epoch": 4.628084179970973,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003046554813203101,
      "loss": 4.7328,
      "mean_token_accuracy": 0.23185273855924607,
      "num_tokens": 105909541.0,
      "step": 51020
    },
    {
      "entropy": 5.496246290206909,
      "epoch": 4.628537735849057,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003046234068699457,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21776219457387924,
      "num_tokens": 105919951.0,
      "step": 51025
    },
    {
      "entropy": 5.470881557464599,
      "epoch": 4.628991291727141,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003045913318069616,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22563156485557556,
      "num_tokens": 105929048.0,
      "step": 51030
    },
    {
      "entropy": 5.488350772857666,
      "epoch": 4.629444847605225,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003045592561320213,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22444267719984054,
      "num_tokens": 105938105.0,
      "step": 51035
    },
    {
      "entropy": 5.487773466110229,
      "epoch": 4.629898403483309,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000304527179845788,
      "loss": 4.934,
      "mean_token_accuracy": 0.21721625328063965,
      "num_tokens": 105948167.0,
      "step": 51040
    },
    {
      "entropy": 5.541027355194092,
      "epoch": 4.630351959361393,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003044951029489252,
      "loss": 4.973,
      "mean_token_accuracy": 0.2184566929936409,
      "num_tokens": 105959387.0,
      "step": 51045
    },
    {
      "entropy": 5.388914346694946,
      "epoch": 4.630805515239477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000304463025442096,
      "loss": 4.8265,
      "mean_token_accuracy": 0.22782854437828065,
      "num_tokens": 105969436.0,
      "step": 51050
    },
    {
      "entropy": 5.511155462265014,
      "epoch": 4.631259071117562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003044309473259641,
      "loss": 4.9476,
      "mean_token_accuracy": 0.22700330913066863,
      "num_tokens": 105979102.0,
      "step": 51055
    },
    {
      "entropy": 5.425132846832275,
      "epoch": 4.631712626995646,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030439886860119265,
      "loss": 4.8166,
      "mean_token_accuracy": 0.2335343211889267,
      "num_tokens": 105989796.0,
      "step": 51060
    },
    {
      "entropy": 5.497353506088257,
      "epoch": 4.63216618287373,
      "grad_norm": 0.984375,
      "learning_rate": 0.000304366789268445,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21967892199754716,
      "num_tokens": 106000194.0,
      "step": 51065
    },
    {
      "entropy": 5.482011318206787,
      "epoch": 4.632619738751814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003043347093283848,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22436463534832002,
      "num_tokens": 106009108.0,
      "step": 51070
    },
    {
      "entropy": 5.431289625167847,
      "epoch": 4.633073294629899,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030430262878167533,
      "loss": 4.8701,
      "mean_token_accuracy": 0.223791141808033,
      "num_tokens": 106019342.0,
      "step": 51075
    },
    {
      "entropy": 5.55504035949707,
      "epoch": 4.633526850507983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003042705476289799,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21051115393638611,
      "num_tokens": 106029079.0,
      "step": 51080
    },
    {
      "entropy": 5.456304550170898,
      "epoch": 4.633980406386067,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003042384658709622,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23263860642910003,
      "num_tokens": 106039406.0,
      "step": 51085
    },
    {
      "entropy": 5.497373437881469,
      "epoch": 4.634433962264151,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003042063835082856,
      "loss": 4.9256,
      "mean_token_accuracy": 0.21883036196231842,
      "num_tokens": 106049603.0,
      "step": 51090
    },
    {
      "entropy": 5.47182879447937,
      "epoch": 4.634887518142235,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003041743005416135,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2277626320719719,
      "num_tokens": 106058692.0,
      "step": 51095
    },
    {
      "entropy": 5.420281314849854,
      "epoch": 4.635341074020319,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003041422169716094,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22817521244287492,
      "num_tokens": 106069307.0,
      "step": 51100
    },
    {
      "entropy": 5.548502111434937,
      "epoch": 4.635794629898403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003041101327989368,
      "loss": 4.9638,
      "mean_token_accuracy": 0.22119733840227127,
      "num_tokens": 106080831.0,
      "step": 51105
    },
    {
      "entropy": 5.4733504295349125,
      "epoch": 4.636248185776488,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003040780480242593,
      "loss": 4.9472,
      "mean_token_accuracy": 0.22077513188123704,
      "num_tokens": 106090007.0,
      "step": 51110
    },
    {
      "entropy": 5.488194561004638,
      "epoch": 4.636701741654572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030404596264824033,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22385130673646927,
      "num_tokens": 106099790.0,
      "step": 51115
    },
    {
      "entropy": 5.460398769378662,
      "epoch": 4.637155297532656,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030401387667154347,
      "loss": 4.8667,
      "mean_token_accuracy": 0.23033728301525117,
      "num_tokens": 106110364.0,
      "step": 51120
    },
    {
      "entropy": 5.392711782455445,
      "epoch": 4.6376088534107405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003039817900948322,
      "loss": 4.816,
      "mean_token_accuracy": 0.22564171105623246,
      "num_tokens": 106121353.0,
      "step": 51125
    },
    {
      "entropy": 5.460520553588867,
      "epoch": 4.638062409288825,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003039497029187702,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21818770915269853,
      "num_tokens": 106130750.0,
      "step": 51130
    },
    {
      "entropy": 5.5143825054168705,
      "epoch": 4.638515965166908,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030391761514402096,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22550010830163955,
      "num_tokens": 106142014.0,
      "step": 51135
    },
    {
      "entropy": 5.490921497344971,
      "epoch": 4.638969521044992,
      "grad_norm": 0.984375,
      "learning_rate": 0.00030388552677124804,
      "loss": 4.919,
      "mean_token_accuracy": 0.22638082057237624,
      "num_tokens": 106152419.0,
      "step": 51140
    },
    {
      "entropy": 5.451019048690796,
      "epoch": 4.639423076923077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000303853437801115,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22412639409303664,
      "num_tokens": 106163365.0,
      "step": 51145
    },
    {
      "entropy": 5.405434799194336,
      "epoch": 4.639876632801161,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003038213482342856,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23140524178743363,
      "num_tokens": 106173324.0,
      "step": 51150
    },
    {
      "entropy": 5.527689838409424,
      "epoch": 4.640330188679245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030378925807142334,
      "loss": 5.0208,
      "mean_token_accuracy": 0.2212878569960594,
      "num_tokens": 106183427.0,
      "step": 51155
    },
    {
      "entropy": 5.521989965438843,
      "epoch": 4.6407837445573294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030375716731319193,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21360668390989304,
      "num_tokens": 106192894.0,
      "step": 51160
    },
    {
      "entropy": 5.4542809009552,
      "epoch": 4.641237300435414,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030372507596025493,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21670289486646652,
      "num_tokens": 106203185.0,
      "step": 51165
    },
    {
      "entropy": 5.59920334815979,
      "epoch": 4.641690856313498,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000303692984013276,
      "loss": 5.0295,
      "mean_token_accuracy": 0.2127094015479088,
      "num_tokens": 106213817.0,
      "step": 51170
    },
    {
      "entropy": 5.442920875549317,
      "epoch": 4.642144412191582,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003036608914729189,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2254399761557579,
      "num_tokens": 106224367.0,
      "step": 51175
    },
    {
      "entropy": 5.3643004417419435,
      "epoch": 4.6425979680696665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003036287983398473,
      "loss": 4.832,
      "mean_token_accuracy": 0.22609135508537292,
      "num_tokens": 106234146.0,
      "step": 51180
    },
    {
      "entropy": 5.5254645347595215,
      "epoch": 4.643051523947751,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030359670461472475,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21191524863243102,
      "num_tokens": 106244910.0,
      "step": 51185
    },
    {
      "entropy": 5.5651243209838865,
      "epoch": 4.643505079825834,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030356461029821516,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20591746717691423,
      "num_tokens": 106254267.0,
      "step": 51190
    },
    {
      "entropy": 5.533767414093018,
      "epoch": 4.643958635703918,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003035325153909821,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21648117452859877,
      "num_tokens": 106264755.0,
      "step": 51195
    },
    {
      "entropy": 5.443782663345337,
      "epoch": 4.644412191582003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003035004198936894,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2180163398385048,
      "num_tokens": 106273964.0,
      "step": 51200
    },
    {
      "entropy": 5.5308915138244625,
      "epoch": 4.644865747460087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003034683238070008,
      "loss": 4.9665,
      "mean_token_accuracy": 0.22099502086639405,
      "num_tokens": 106283933.0,
      "step": 51205
    },
    {
      "entropy": 5.5161717414855955,
      "epoch": 4.645319303338171,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00030343622713157985,
      "loss": 4.905,
      "mean_token_accuracy": 0.2222788393497467,
      "num_tokens": 106294075.0,
      "step": 51210
    },
    {
      "entropy": 5.42706789970398,
      "epoch": 4.6457728592162555,
      "grad_norm": 1.25,
      "learning_rate": 0.0003034041298680906,
      "loss": 4.9062,
      "mean_token_accuracy": 0.22350094020366668,
      "num_tokens": 106304970.0,
      "step": 51215
    },
    {
      "entropy": 5.461996459960938,
      "epoch": 4.64622641509434,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003033720320171966,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21656594276428223,
      "num_tokens": 106315610.0,
      "step": 51220
    },
    {
      "entropy": 5.5254011154174805,
      "epoch": 4.646679970972424,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003033399335795619,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21658933460712432,
      "num_tokens": 106326302.0,
      "step": 51225
    },
    {
      "entropy": 5.500261068344116,
      "epoch": 4.647133526850508,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030330783455585,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22024667710065843,
      "num_tokens": 106336350.0,
      "step": 51230
    },
    {
      "entropy": 5.4743537425994875,
      "epoch": 4.647587082728592,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003032757349467249,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21386634409427643,
      "num_tokens": 106348233.0,
      "step": 51235
    },
    {
      "entropy": 5.435495853424072,
      "epoch": 4.648040638606677,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030324363475285046,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22019711136817932,
      "num_tokens": 106357992.0,
      "step": 51240
    },
    {
      "entropy": 5.461936521530151,
      "epoch": 4.64849419448476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003032115339748904,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22215496003627777,
      "num_tokens": 106367683.0,
      "step": 51245
    },
    {
      "entropy": 5.478367853164673,
      "epoch": 4.6489477503628445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003031794326135086,
      "loss": 4.8961,
      "mean_token_accuracy": 0.2247927963733673,
      "num_tokens": 106378275.0,
      "step": 51250
    },
    {
      "entropy": 5.483480548858642,
      "epoch": 4.649401306240929,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030314733066936903,
      "loss": 4.928,
      "mean_token_accuracy": 0.22304464280605316,
      "num_tokens": 106387811.0,
      "step": 51255
    },
    {
      "entropy": 5.45595154762268,
      "epoch": 4.649854862119013,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003031152281431355,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21518058627843856,
      "num_tokens": 106398327.0,
      "step": 51260
    },
    {
      "entropy": 5.436999845504761,
      "epoch": 4.650308417997097,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030308312503547183,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2207299828529358,
      "num_tokens": 106408602.0,
      "step": 51265
    },
    {
      "entropy": 5.460390186309814,
      "epoch": 4.6507619738751815,
      "grad_norm": 1.171875,
      "learning_rate": 0.000303051021347042,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22798873335123063,
      "num_tokens": 106418717.0,
      "step": 51270
    },
    {
      "entropy": 5.419030427932739,
      "epoch": 4.651215529753266,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030301891707850987,
      "loss": 4.7705,
      "mean_token_accuracy": 0.23062815070152282,
      "num_tokens": 106428311.0,
      "step": 51275
    },
    {
      "entropy": 5.488641929626465,
      "epoch": 4.65166908563135,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003029868122305395,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21636989563703538,
      "num_tokens": 106438772.0,
      "step": 51280
    },
    {
      "entropy": 5.511090421676636,
      "epoch": 4.652122641509434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030295470680379456,
      "loss": 4.9126,
      "mean_token_accuracy": 0.22514560222625732,
      "num_tokens": 106448895.0,
      "step": 51285
    },
    {
      "entropy": 5.470692873001099,
      "epoch": 4.652576197387518,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003029226007989393,
      "loss": 4.9886,
      "mean_token_accuracy": 0.22270452976226807,
      "num_tokens": 106459711.0,
      "step": 51290
    },
    {
      "entropy": 5.496733474731445,
      "epoch": 4.653029753265602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030289049421663744,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22600816190242767,
      "num_tokens": 106469691.0,
      "step": 51295
    },
    {
      "entropy": 5.486766290664673,
      "epoch": 4.653483309143686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003028583870575531,
      "loss": 4.912,
      "mean_token_accuracy": 0.21905265897512435,
      "num_tokens": 106479894.0,
      "step": 51300
    },
    {
      "entropy": 5.466835832595825,
      "epoch": 4.6539368650217705,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003028262793223502,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22809371799230577,
      "num_tokens": 106491444.0,
      "step": 51305
    },
    {
      "entropy": 5.487726354598999,
      "epoch": 4.654390420899855,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030279417101169276,
      "loss": 4.9202,
      "mean_token_accuracy": 0.22329962402582168,
      "num_tokens": 106501633.0,
      "step": 51310
    },
    {
      "entropy": 5.4619182586669925,
      "epoch": 4.654843976777939,
      "grad_norm": 1.125,
      "learning_rate": 0.0003027620621262448,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22639932334423066,
      "num_tokens": 106511272.0,
      "step": 51315
    },
    {
      "entropy": 5.529841661453247,
      "epoch": 4.655297532656023,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003027299526666703,
      "loss": 5.0406,
      "mean_token_accuracy": 0.21713468581438064,
      "num_tokens": 106521977.0,
      "step": 51320
    },
    {
      "entropy": 5.4933562755584715,
      "epoch": 4.655751088534108,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030269784263363327,
      "loss": 4.9554,
      "mean_token_accuracy": 0.21890002340078354,
      "num_tokens": 106532802.0,
      "step": 51325
    },
    {
      "entropy": 5.438062429428101,
      "epoch": 4.656204644412192,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030266573202779784,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22182261645793916,
      "num_tokens": 106542249.0,
      "step": 51330
    },
    {
      "entropy": 5.485075759887695,
      "epoch": 4.656658200290276,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030263362084982805,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22300464808940887,
      "num_tokens": 106552501.0,
      "step": 51335
    },
    {
      "entropy": 5.4531426429748535,
      "epoch": 4.65711175616836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030260150910038793,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2313928186893463,
      "num_tokens": 106561901.0,
      "step": 51340
    },
    {
      "entropy": 5.44431734085083,
      "epoch": 4.657565312046444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030256939678014155,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22411516904830933,
      "num_tokens": 106572104.0,
      "step": 51345
    },
    {
      "entropy": 5.485248327255249,
      "epoch": 4.658018867924528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003025372838897529,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2162074401974678,
      "num_tokens": 106583436.0,
      "step": 51350
    },
    {
      "entropy": 5.520272159576416,
      "epoch": 4.658472423802612,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003025051704298864,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21391209214925766,
      "num_tokens": 106593879.0,
      "step": 51355
    },
    {
      "entropy": 5.501303100585938,
      "epoch": 4.6589259796806965,
      "grad_norm": 0.96875,
      "learning_rate": 0.00030247305640120585,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21389725506305696,
      "num_tokens": 106604774.0,
      "step": 51360
    },
    {
      "entropy": 5.5436413288116455,
      "epoch": 4.659379535558781,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003024409418043756,
      "loss": 5.0069,
      "mean_token_accuracy": 0.2119535446166992,
      "num_tokens": 106616125.0,
      "step": 51365
    },
    {
      "entropy": 5.527444553375244,
      "epoch": 4.659833091436865,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003024088266400595,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21912021934986115,
      "num_tokens": 106625171.0,
      "step": 51370
    },
    {
      "entropy": 5.450742483139038,
      "epoch": 4.660286647314949,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030237671090892204,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21912358105182647,
      "num_tokens": 106636455.0,
      "step": 51375
    },
    {
      "entropy": 5.472281169891358,
      "epoch": 4.660740203193034,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003023445946116272,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2166170135140419,
      "num_tokens": 106645956.0,
      "step": 51380
    },
    {
      "entropy": 5.514968538284302,
      "epoch": 4.661193759071118,
      "grad_norm": 0.96875,
      "learning_rate": 0.00030231247774883913,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2182717889547348,
      "num_tokens": 106655562.0,
      "step": 51385
    },
    {
      "entropy": 5.51924319267273,
      "epoch": 4.661647314949201,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003022803603212221,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21049704849720002,
      "num_tokens": 106666620.0,
      "step": 51390
    },
    {
      "entropy": 5.477151679992676,
      "epoch": 4.662100870827286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003022482423294403,
      "loss": 4.91,
      "mean_token_accuracy": 0.21944237500429153,
      "num_tokens": 106676881.0,
      "step": 51395
    },
    {
      "entropy": 5.49418683052063,
      "epoch": 4.66255442670537,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003022161237741578,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22313937395811081,
      "num_tokens": 106687765.0,
      "step": 51400
    },
    {
      "entropy": 5.455647373199463,
      "epoch": 4.663007982583454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030218400465603895,
      "loss": 4.906,
      "mean_token_accuracy": 0.22043117135763168,
      "num_tokens": 106698796.0,
      "step": 51405
    },
    {
      "entropy": 5.489095163345337,
      "epoch": 4.663461538461538,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000302151884975748,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2226927772164345,
      "num_tokens": 106709439.0,
      "step": 51410
    },
    {
      "entropy": 5.532384061813355,
      "epoch": 4.663915094339623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003021197647339492,
      "loss": 5.0215,
      "mean_token_accuracy": 0.21464200615882872,
      "num_tokens": 106720325.0,
      "step": 51415
    },
    {
      "entropy": 5.543565940856934,
      "epoch": 4.664368650217707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003020876439313066,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2215721860527992,
      "num_tokens": 106730526.0,
      "step": 51420
    },
    {
      "entropy": 5.491705417633057,
      "epoch": 4.664822206095791,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030205552256848476,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22172033935785293,
      "num_tokens": 106741142.0,
      "step": 51425
    },
    {
      "entropy": 5.55950779914856,
      "epoch": 4.665275761973875,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030202340064614783,
      "loss": 5.0274,
      "mean_token_accuracy": 0.21627719700336456,
      "num_tokens": 106751843.0,
      "step": 51430
    },
    {
      "entropy": 5.424393129348755,
      "epoch": 4.66572931785196,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003019912781649599,
      "loss": 4.7907,
      "mean_token_accuracy": 0.2287256896495819,
      "num_tokens": 106761726.0,
      "step": 51435
    },
    {
      "entropy": 5.464760065078735,
      "epoch": 4.666182873730044,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00030195915512558564,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22485509514808655,
      "num_tokens": 106771829.0,
      "step": 51440
    },
    {
      "entropy": 5.391293048858643,
      "epoch": 4.666636429608127,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003019270315286892,
      "loss": 4.8465,
      "mean_token_accuracy": 0.22231278717517852,
      "num_tokens": 106783457.0,
      "step": 51445
    },
    {
      "entropy": 5.4620506286621096,
      "epoch": 4.667089985486212,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030189490737493477,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22789919674396514,
      "num_tokens": 106793290.0,
      "step": 51450
    },
    {
      "entropy": 5.479299449920655,
      "epoch": 4.667543541364296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030186278266498677,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22044607400894164,
      "num_tokens": 106804412.0,
      "step": 51455
    },
    {
      "entropy": 5.547525691986084,
      "epoch": 4.66799709724238,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003018306573995097,
      "loss": 5.0272,
      "mean_token_accuracy": 0.21893570572137833,
      "num_tokens": 106815353.0,
      "step": 51460
    },
    {
      "entropy": 5.490588045120239,
      "epoch": 4.668450653120464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030179853157916777,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22079442888498307,
      "num_tokens": 106826350.0,
      "step": 51465
    },
    {
      "entropy": 5.527143955230713,
      "epoch": 4.668904208998549,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003017664052046253,
      "loss": 4.9521,
      "mean_token_accuracy": 0.222306028008461,
      "num_tokens": 106837434.0,
      "step": 51470
    },
    {
      "entropy": 5.484354448318482,
      "epoch": 4.669357764876633,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030173427827654683,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21549189537763597,
      "num_tokens": 106847829.0,
      "step": 51475
    },
    {
      "entropy": 5.59851484298706,
      "epoch": 4.669811320754717,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003017021507955967,
      "loss": 5.0425,
      "mean_token_accuracy": 0.21087829470634462,
      "num_tokens": 106858519.0,
      "step": 51480
    },
    {
      "entropy": 5.501718378067016,
      "epoch": 4.670264876632801,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030167002276243915,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2247650682926178,
      "num_tokens": 106868614.0,
      "step": 51485
    },
    {
      "entropy": 5.521900463104248,
      "epoch": 4.670718432510886,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030163789417773885,
      "loss": 5.0513,
      "mean_token_accuracy": 0.21387587785720824,
      "num_tokens": 106879592.0,
      "step": 51490
    },
    {
      "entropy": 5.518009853363037,
      "epoch": 4.67117198838897,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003016057650421601,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2172401413321495,
      "num_tokens": 106890083.0,
      "step": 51495
    },
    {
      "entropy": 5.5422625064849855,
      "epoch": 4.671625544267053,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030157363535636726,
      "loss": 4.9997,
      "mean_token_accuracy": 0.21998457312583924,
      "num_tokens": 106900672.0,
      "step": 51500
    },
    {
      "entropy": 5.483010959625244,
      "epoch": 4.672079100145138,
      "grad_norm": 1.078125,
      "learning_rate": 0.000301541505121025,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22338854223489762,
      "num_tokens": 106910396.0,
      "step": 51505
    },
    {
      "entropy": 5.503655004501343,
      "epoch": 4.672532656023222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003015093743367976,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21638180762529374,
      "num_tokens": 106919742.0,
      "step": 51510
    },
    {
      "entropy": 5.445378637313842,
      "epoch": 4.672986211901306,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003014772430043495,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2309053659439087,
      "num_tokens": 106930837.0,
      "step": 51515
    },
    {
      "entropy": 5.46974811553955,
      "epoch": 4.67343976777939,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030144511112434536,
      "loss": 4.8919,
      "mean_token_accuracy": 0.2261481150984764,
      "num_tokens": 106942023.0,
      "step": 51520
    },
    {
      "entropy": 5.511404132843017,
      "epoch": 4.673893323657475,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030141297869744956,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21489516347646714,
      "num_tokens": 106951374.0,
      "step": 51525
    },
    {
      "entropy": 5.571134805679321,
      "epoch": 4.674346879535559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030138084572432663,
      "loss": 5.0019,
      "mean_token_accuracy": 0.21091508269309997,
      "num_tokens": 106961714.0,
      "step": 51530
    },
    {
      "entropy": 5.493590402603149,
      "epoch": 4.674800435413643,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003013487122056411,
      "loss": 4.9197,
      "mean_token_accuracy": 0.22348572313785553,
      "num_tokens": 106971783.0,
      "step": 51535
    },
    {
      "entropy": 5.428099012374878,
      "epoch": 4.6752539912917275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003013165781420575,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2296299085021019,
      "num_tokens": 106982047.0,
      "step": 51540
    },
    {
      "entropy": 5.467019033432007,
      "epoch": 4.675707547169811,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030128444353424034,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22423485070466995,
      "num_tokens": 106993063.0,
      "step": 51545
    },
    {
      "entropy": 5.476859092712402,
      "epoch": 4.676161103047895,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030125230838285415,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21622429341077803,
      "num_tokens": 107002865.0,
      "step": 51550
    },
    {
      "entropy": 5.555632066726685,
      "epoch": 4.676614658925979,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030122017268856353,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22392915487289428,
      "num_tokens": 107013007.0,
      "step": 51555
    },
    {
      "entropy": 5.5482001304626465,
      "epoch": 4.677068214804064,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030118803645203315,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21638843268156052,
      "num_tokens": 107023109.0,
      "step": 51560
    },
    {
      "entropy": 5.447467947006226,
      "epoch": 4.677521770682148,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003011558996739274,
      "loss": 4.881,
      "mean_token_accuracy": 0.22342034727334975,
      "num_tokens": 107033824.0,
      "step": 51565
    },
    {
      "entropy": 5.495787858963013,
      "epoch": 4.677975326560232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000301123762354911,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22064177244901656,
      "num_tokens": 107043529.0,
      "step": 51570
    },
    {
      "entropy": 5.527210760116577,
      "epoch": 4.6784288824383164,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003010916244956486,
      "loss": 4.9714,
      "mean_token_accuracy": 0.22642166763544083,
      "num_tokens": 107054378.0,
      "step": 51575
    },
    {
      "entropy": 5.47856502532959,
      "epoch": 4.678882438316401,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003010594860968048,
      "loss": 4.839,
      "mean_token_accuracy": 0.21959611922502517,
      "num_tokens": 107064300.0,
      "step": 51580
    },
    {
      "entropy": 5.528578186035157,
      "epoch": 4.679335994194485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000301027347159044,
      "loss": 4.9741,
      "mean_token_accuracy": 0.21612943708896637,
      "num_tokens": 107074560.0,
      "step": 51585
    },
    {
      "entropy": 5.481158113479614,
      "epoch": 4.679789550072569,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003009952076830312,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2143137812614441,
      "num_tokens": 107084574.0,
      "step": 51590
    },
    {
      "entropy": 5.45859375,
      "epoch": 4.6802431059506535,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003009630676694308,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2158583104610443,
      "num_tokens": 107095207.0,
      "step": 51595
    },
    {
      "entropy": 5.48070011138916,
      "epoch": 4.680696661828737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030093092711890756,
      "loss": 4.91,
      "mean_token_accuracy": 0.2213663011789322,
      "num_tokens": 107106091.0,
      "step": 51600
    },
    {
      "entropy": 5.502422428131103,
      "epoch": 4.681150217706821,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003008987860321261,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21382594257593154,
      "num_tokens": 107117753.0,
      "step": 51605
    },
    {
      "entropy": 5.445098733901977,
      "epoch": 4.681603773584905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003008666444097512,
      "loss": 4.8452,
      "mean_token_accuracy": 0.23062461614608765,
      "num_tokens": 107128131.0,
      "step": 51610
    },
    {
      "entropy": 5.574739837646485,
      "epoch": 4.68205732946299,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003008345022524475,
      "loss": 5.0728,
      "mean_token_accuracy": 0.20365942269563675,
      "num_tokens": 107139447.0,
      "step": 51615
    },
    {
      "entropy": 5.4971319198608395,
      "epoch": 4.682510885341074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030080235956087963,
      "loss": 4.94,
      "mean_token_accuracy": 0.21670219898223878,
      "num_tokens": 107149721.0,
      "step": 51620
    },
    {
      "entropy": 5.501498126983643,
      "epoch": 4.682964441219158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003007702163357125,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21637229174375533,
      "num_tokens": 107159814.0,
      "step": 51625
    },
    {
      "entropy": 5.461524915695191,
      "epoch": 4.6834179970972425,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030073807257761074,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22118045538663864,
      "num_tokens": 107169012.0,
      "step": 51630
    },
    {
      "entropy": 5.513629579544068,
      "epoch": 4.683871552975327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030070592828723903,
      "loss": 4.9521,
      "mean_token_accuracy": 0.213104647397995,
      "num_tokens": 107179571.0,
      "step": 51635
    },
    {
      "entropy": 5.513889932632447,
      "epoch": 4.684325108853411,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003006737834652621,
      "loss": 4.9192,
      "mean_token_accuracy": 0.22223443537950516,
      "num_tokens": 107190535.0,
      "step": 51640
    },
    {
      "entropy": 5.536368703842163,
      "epoch": 4.684778664731495,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003006416381123449,
      "loss": 4.9349,
      "mean_token_accuracy": 0.2285557746887207,
      "num_tokens": 107200361.0,
      "step": 51645
    },
    {
      "entropy": 5.501131153106689,
      "epoch": 4.68523222060958,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030060949222915205,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21663871109485627,
      "num_tokens": 107211080.0,
      "step": 51650
    },
    {
      "entropy": 5.521431827545166,
      "epoch": 4.685685776487663,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003005773458163484,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21676278561353685,
      "num_tokens": 107220556.0,
      "step": 51655
    },
    {
      "entropy": 5.520729923248291,
      "epoch": 4.686139332365747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030054519887459864,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21703554391860963,
      "num_tokens": 107230693.0,
      "step": 51660
    },
    {
      "entropy": 5.514331388473511,
      "epoch": 4.6865928882438315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030051305140456775,
      "loss": 4.9763,
      "mean_token_accuracy": 0.21421052068471907,
      "num_tokens": 107240926.0,
      "step": 51665
    },
    {
      "entropy": 5.475758123397827,
      "epoch": 4.687046444121916,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030048090340692034,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21628712564706803,
      "num_tokens": 107251425.0,
      "step": 51670
    },
    {
      "entropy": 5.4976362705230715,
      "epoch": 4.6875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030044875488232145,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22380864173173903,
      "num_tokens": 107262046.0,
      "step": 51675
    },
    {
      "entropy": 5.440053510665893,
      "epoch": 4.687953555878084,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003004166058314358,
      "loss": 4.885,
      "mean_token_accuracy": 0.22940307408571242,
      "num_tokens": 107273141.0,
      "step": 51680
    },
    {
      "entropy": 5.516154670715332,
      "epoch": 4.6884071117561685,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003003844562549282,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21975751370191574,
      "num_tokens": 107282497.0,
      "step": 51685
    },
    {
      "entropy": 5.44742169380188,
      "epoch": 4.688860667634253,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030035230615346373,
      "loss": 4.8599,
      "mean_token_accuracy": 0.23039280325174333,
      "num_tokens": 107292710.0,
      "step": 51690
    },
    {
      "entropy": 5.40895447731018,
      "epoch": 4.689314223512337,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030032015552770695,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23286664485931396,
      "num_tokens": 107303157.0,
      "step": 51695
    },
    {
      "entropy": 5.4735517501831055,
      "epoch": 4.68976777939042,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030028800437832296,
      "loss": 4.933,
      "mean_token_accuracy": 0.2171658307313919,
      "num_tokens": 107313503.0,
      "step": 51700
    },
    {
      "entropy": 5.452723360061645,
      "epoch": 4.690221335268505,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003002558527059766,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22652594298124312,
      "num_tokens": 107322276.0,
      "step": 51705
    },
    {
      "entropy": 5.4737975120544435,
      "epoch": 4.690674891146589,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003002237005113328,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21663490533828736,
      "num_tokens": 107332876.0,
      "step": 51710
    },
    {
      "entropy": 5.500940036773682,
      "epoch": 4.691128447024673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030019154779505644,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22083351016044617,
      "num_tokens": 107343262.0,
      "step": 51715
    },
    {
      "entropy": 5.458014583587646,
      "epoch": 4.6915820029027575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003001593945578124,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21551592648029327,
      "num_tokens": 107354281.0,
      "step": 51720
    },
    {
      "entropy": 5.401322221755981,
      "epoch": 4.692035558780842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003001272408002657,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22808231562376022,
      "num_tokens": 107364011.0,
      "step": 51725
    },
    {
      "entropy": 5.428802537918091,
      "epoch": 4.692489114658926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030009508652308124,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22148145437240602,
      "num_tokens": 107372943.0,
      "step": 51730
    },
    {
      "entropy": 5.459691286087036,
      "epoch": 4.69294267053701,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030006293172692405,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22410086542367935,
      "num_tokens": 107382838.0,
      "step": 51735
    },
    {
      "entropy": 5.541161775588989,
      "epoch": 4.693396226415095,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030003077641245897,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21990411579608918,
      "num_tokens": 107393250.0,
      "step": 51740
    },
    {
      "entropy": 5.507727336883545,
      "epoch": 4.693849782293179,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029999862058035116,
      "loss": 4.812,
      "mean_token_accuracy": 0.2281745344400406,
      "num_tokens": 107403153.0,
      "step": 51745
    },
    {
      "entropy": 5.473035717010498,
      "epoch": 4.694303338171263,
      "grad_norm": 1.0,
      "learning_rate": 0.00029996646423126544,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21921346187591553,
      "num_tokens": 107413092.0,
      "step": 51750
    },
    {
      "entropy": 5.516891098022461,
      "epoch": 4.6947568940493465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002999343073658669,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21589196622371673,
      "num_tokens": 107423561.0,
      "step": 51755
    },
    {
      "entropy": 5.538142395019531,
      "epoch": 4.695210449927431,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029990214998482046,
      "loss": 4.9551,
      "mean_token_accuracy": 0.2212556555867195,
      "num_tokens": 107433954.0,
      "step": 51760
    },
    {
      "entropy": 5.502284860610962,
      "epoch": 4.695664005805515,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002998699920887913,
      "loss": 4.87,
      "mean_token_accuracy": 0.22447915822267533,
      "num_tokens": 107443150.0,
      "step": 51765
    },
    {
      "entropy": 5.4500843524932865,
      "epoch": 4.696117561683599,
      "grad_norm": 1.25,
      "learning_rate": 0.0002998378336784443,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22154585272073746,
      "num_tokens": 107454250.0,
      "step": 51770
    },
    {
      "entropy": 5.377288961410523,
      "epoch": 4.6965711175616836,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002998056747544446,
      "loss": 4.8558,
      "mean_token_accuracy": 0.23081037998199463,
      "num_tokens": 107465058.0,
      "step": 51775
    },
    {
      "entropy": 5.424403953552246,
      "epoch": 4.697024673439768,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002997735153174572,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2280406668782234,
      "num_tokens": 107475449.0,
      "step": 51780
    },
    {
      "entropy": 5.634197950363159,
      "epoch": 4.697478229317852,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029974135536814725,
      "loss": 5.1076,
      "mean_token_accuracy": 0.21342390179634094,
      "num_tokens": 107486645.0,
      "step": 51785
    },
    {
      "entropy": 5.490660333633423,
      "epoch": 4.697931785195936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002997091949071797,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22062095403671264,
      "num_tokens": 107497242.0,
      "step": 51790
    },
    {
      "entropy": 5.539371061325073,
      "epoch": 4.698385341074021,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002996770339352197,
      "loss": 4.9488,
      "mean_token_accuracy": 0.22294629663228988,
      "num_tokens": 107507596.0,
      "step": 51795
    },
    {
      "entropy": 5.550246953964233,
      "epoch": 4.698838896952105,
      "grad_norm": 1.125,
      "learning_rate": 0.0002996448724529324,
      "loss": 5.018,
      "mean_token_accuracy": 0.21529005467891693,
      "num_tokens": 107517795.0,
      "step": 51800
    },
    {
      "entropy": 5.519339895248413,
      "epoch": 4.699292452830189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029961271046098284,
      "loss": 4.8981,
      "mean_token_accuracy": 0.2233282059431076,
      "num_tokens": 107527805.0,
      "step": 51805
    },
    {
      "entropy": 5.483393716812134,
      "epoch": 4.6997460087082725,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029958054796003615,
      "loss": 4.9181,
      "mean_token_accuracy": 0.22368668466806413,
      "num_tokens": 107537396.0,
      "step": 51810
    },
    {
      "entropy": 5.434419012069702,
      "epoch": 4.700199564586357,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002995483849507575,
      "loss": 4.9014,
      "mean_token_accuracy": 0.22151425778865813,
      "num_tokens": 107546876.0,
      "step": 51815
    },
    {
      "entropy": 5.467576694488526,
      "epoch": 4.700653120464441,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000299516221433812,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2264159470796585,
      "num_tokens": 107556865.0,
      "step": 51820
    },
    {
      "entropy": 5.480198907852173,
      "epoch": 4.701106676342525,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029948405740986466,
      "loss": 4.913,
      "mean_token_accuracy": 0.21806277334690094,
      "num_tokens": 107566305.0,
      "step": 51825
    },
    {
      "entropy": 5.573738145828247,
      "epoch": 4.70156023222061,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002994518928795809,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21838660836219786,
      "num_tokens": 107575640.0,
      "step": 51830
    },
    {
      "entropy": 5.536039161682129,
      "epoch": 4.702013788098694,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029941972784362574,
      "loss": 4.9923,
      "mean_token_accuracy": 0.2212309718132019,
      "num_tokens": 107585955.0,
      "step": 51835
    },
    {
      "entropy": 5.514097118377686,
      "epoch": 4.702467343976778,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029938756230266436,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2154555007815361,
      "num_tokens": 107596911.0,
      "step": 51840
    },
    {
      "entropy": 5.4345252990722654,
      "epoch": 4.702920899854862,
      "grad_norm": 1.109375,
      "learning_rate": 0.000299355396257362,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22628349363803862,
      "num_tokens": 107606117.0,
      "step": 51845
    },
    {
      "entropy": 5.417080545425415,
      "epoch": 4.703374455732947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002993232297083838,
      "loss": 4.7717,
      "mean_token_accuracy": 0.24208221435546876,
      "num_tokens": 107616363.0,
      "step": 51850
    },
    {
      "entropy": 5.422995471954346,
      "epoch": 4.70382801161103,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002992910626563951,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2234540805220604,
      "num_tokens": 107627402.0,
      "step": 51855
    },
    {
      "entropy": 5.444747686386108,
      "epoch": 4.704281567489114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029925889510206094,
      "loss": 4.913,
      "mean_token_accuracy": 0.22336110472679138,
      "num_tokens": 107637446.0,
      "step": 51860
    },
    {
      "entropy": 5.4411036491394045,
      "epoch": 4.704735123367199,
      "grad_norm": 1.0,
      "learning_rate": 0.0002992267270460467,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22295713126659394,
      "num_tokens": 107647386.0,
      "step": 51865
    },
    {
      "entropy": 5.54599781036377,
      "epoch": 4.705188679245283,
      "grad_norm": 1.125,
      "learning_rate": 0.0002991945584890175,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2243222028017044,
      "num_tokens": 107657881.0,
      "step": 51870
    },
    {
      "entropy": 5.549424076080323,
      "epoch": 4.705642235123367,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029916238943163874,
      "loss": 4.926,
      "mean_token_accuracy": 0.22376393526792526,
      "num_tokens": 107668478.0,
      "step": 51875
    },
    {
      "entropy": 5.574392652511596,
      "epoch": 4.706095791001451,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002991302198745756,
      "loss": 5.0601,
      "mean_token_accuracy": 0.21699783504009246,
      "num_tokens": 107679149.0,
      "step": 51880
    },
    {
      "entropy": 5.522529935836792,
      "epoch": 4.706549346879536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002990980498184934,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2125596970319748,
      "num_tokens": 107689471.0,
      "step": 51885
    },
    {
      "entropy": 5.4733353614807125,
      "epoch": 4.70700290275762,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029906587926405734,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2262137085199356,
      "num_tokens": 107700604.0,
      "step": 51890
    },
    {
      "entropy": 5.3841094970703125,
      "epoch": 4.707456458635704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002990337082119327,
      "loss": 4.833,
      "mean_token_accuracy": 0.2235601931810379,
      "num_tokens": 107710268.0,
      "step": 51895
    },
    {
      "entropy": 5.509324979782105,
      "epoch": 4.707910014513788,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029900153666278503,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2182037875056267,
      "num_tokens": 107720328.0,
      "step": 51900
    },
    {
      "entropy": 5.480173063278198,
      "epoch": 4.708363570391873,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029896936461727924,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2226466029882431,
      "num_tokens": 107730295.0,
      "step": 51905
    },
    {
      "entropy": 5.452961492538452,
      "epoch": 4.708817126269956,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029893719207608105,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21838720589876176,
      "num_tokens": 107741334.0,
      "step": 51910
    },
    {
      "entropy": 5.459913492202759,
      "epoch": 4.70927068214804,
      "grad_norm": 0.921875,
      "learning_rate": 0.00029890501903985555,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21926838755607606,
      "num_tokens": 107752060.0,
      "step": 51915
    },
    {
      "entropy": 5.4865048885345455,
      "epoch": 4.709724238026125,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029887284550926817,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22390855997800826,
      "num_tokens": 107762188.0,
      "step": 51920
    },
    {
      "entropy": 5.501448774337769,
      "epoch": 4.710177793904209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002988406714849843,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21564466655254363,
      "num_tokens": 107771670.0,
      "step": 51925
    },
    {
      "entropy": 5.481127977371216,
      "epoch": 4.710631349782293,
      "grad_norm": 1.125,
      "learning_rate": 0.0002988084969676692,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22341742366552353,
      "num_tokens": 107781734.0,
      "step": 51930
    },
    {
      "entropy": 5.522968196868897,
      "epoch": 4.711084905660377,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029877632195798843,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21435339450836183,
      "num_tokens": 107792159.0,
      "step": 51935
    },
    {
      "entropy": 5.459611463546753,
      "epoch": 4.711538461538462,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029874414645660713,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2235391229391098,
      "num_tokens": 107803153.0,
      "step": 51940
    },
    {
      "entropy": 5.4081535816192625,
      "epoch": 4.711992017416546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002987119704641909,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2243517741560936,
      "num_tokens": 107813214.0,
      "step": 51945
    },
    {
      "entropy": 5.464072847366333,
      "epoch": 4.71244557329463,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002986797939814051,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22643804401159287,
      "num_tokens": 107823976.0,
      "step": 51950
    },
    {
      "entropy": 5.505411958694458,
      "epoch": 4.712899129172714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029864761700891507,
      "loss": 4.9455,
      "mean_token_accuracy": 0.22247597873210906,
      "num_tokens": 107834127.0,
      "step": 51955
    },
    {
      "entropy": 5.545209074020386,
      "epoch": 4.713352685050799,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029861543954738635,
      "loss": 5.0144,
      "mean_token_accuracy": 0.21233623176813127,
      "num_tokens": 107843753.0,
      "step": 51960
    },
    {
      "entropy": 5.537936496734619,
      "epoch": 4.713806240928882,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029858326159748434,
      "loss": 4.9253,
      "mean_token_accuracy": 0.22475329488515855,
      "num_tokens": 107854690.0,
      "step": 51965
    },
    {
      "entropy": 5.493241882324218,
      "epoch": 4.714259796806966,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029855108315987437,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22429232746362687,
      "num_tokens": 107865235.0,
      "step": 51970
    },
    {
      "entropy": 5.484741497039795,
      "epoch": 4.714713352685051,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000298518904235222,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22358517944812775,
      "num_tokens": 107876052.0,
      "step": 51975
    },
    {
      "entropy": 5.442521953582764,
      "epoch": 4.715166908563135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002984867248241928,
      "loss": 4.9153,
      "mean_token_accuracy": 0.22842399626970292,
      "num_tokens": 107886187.0,
      "step": 51980
    },
    {
      "entropy": 5.495499086380005,
      "epoch": 4.715620464441219,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002984545449274521,
      "loss": 4.942,
      "mean_token_accuracy": 0.22756315171718597,
      "num_tokens": 107896995.0,
      "step": 51985
    },
    {
      "entropy": 5.564328908920288,
      "epoch": 4.7160740203193035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002984223645456655,
      "loss": 4.957,
      "mean_token_accuracy": 0.2203683987259865,
      "num_tokens": 107907868.0,
      "step": 51990
    },
    {
      "entropy": 5.508925485610962,
      "epoch": 4.716527576197388,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029839018367949825,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21824754327535628,
      "num_tokens": 107917564.0,
      "step": 51995
    },
    {
      "entropy": 5.507888507843018,
      "epoch": 4.716981132075472,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002983580023296162,
      "loss": 4.876,
      "mean_token_accuracy": 0.23203889280557632,
      "num_tokens": 107926853.0,
      "step": 52000
    },
    {
      "entropy": 5.453591156005859,
      "epoch": 4.717434687953556,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029832582049668465,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2185967355966568,
      "num_tokens": 107937440.0,
      "step": 52005
    },
    {
      "entropy": 5.524221467971802,
      "epoch": 4.71788824383164,
      "grad_norm": 1.640625,
      "learning_rate": 0.00029829363818136917,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21634705364704132,
      "num_tokens": 107947750.0,
      "step": 52010
    },
    {
      "entropy": 5.550691270828247,
      "epoch": 4.718341799709724,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002982614553843354,
      "loss": 4.9571,
      "mean_token_accuracy": 0.22223757356405258,
      "num_tokens": 107959646.0,
      "step": 52015
    },
    {
      "entropy": 5.528194904327393,
      "epoch": 4.718795355587808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029822927210624873,
      "loss": 4.975,
      "mean_token_accuracy": 0.21532842963933946,
      "num_tokens": 107969678.0,
      "step": 52020
    },
    {
      "entropy": 5.496554279327393,
      "epoch": 4.719248911465892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029819708834777483,
      "loss": 4.9523,
      "mean_token_accuracy": 0.22201904207468032,
      "num_tokens": 107979589.0,
      "step": 52025
    },
    {
      "entropy": 5.420658349990845,
      "epoch": 4.719702467343977,
      "grad_norm": 0.953125,
      "learning_rate": 0.0002981649041095792,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22504721581935883,
      "num_tokens": 107990682.0,
      "step": 52030
    },
    {
      "entropy": 5.493080186843872,
      "epoch": 4.720156023222061,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002981327193923275,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22463301569223404,
      "num_tokens": 108000999.0,
      "step": 52035
    },
    {
      "entropy": 5.434778213500977,
      "epoch": 4.720609579100145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002981005341966852,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22148560881614685,
      "num_tokens": 108010709.0,
      "step": 52040
    },
    {
      "entropy": 5.421860599517823,
      "epoch": 4.7210631349782295,
      "grad_norm": 1.0,
      "learning_rate": 0.000298068348523318,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22718224823474883,
      "num_tokens": 108020522.0,
      "step": 52045
    },
    {
      "entropy": 5.419876432418823,
      "epoch": 4.721516690856314,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002980361623728915,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22988466322422027,
      "num_tokens": 108031387.0,
      "step": 52050
    },
    {
      "entropy": 5.509647989273072,
      "epoch": 4.721970246734398,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002980039757460713,
      "loss": 4.9123,
      "mean_token_accuracy": 0.22044145464897155,
      "num_tokens": 108041653.0,
      "step": 52055
    },
    {
      "entropy": 5.586578035354615,
      "epoch": 4.722423802612482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029797178864352306,
      "loss": 5.0529,
      "mean_token_accuracy": 0.21467031240463258,
      "num_tokens": 108051969.0,
      "step": 52060
    },
    {
      "entropy": 5.4729210376739506,
      "epoch": 4.722877358490566,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029793960106591226,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21854539811611176,
      "num_tokens": 108061364.0,
      "step": 52065
    },
    {
      "entropy": 5.465286064147949,
      "epoch": 4.72333091436865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002979074130139047,
      "loss": 4.9691,
      "mean_token_accuracy": 0.22039879262447357,
      "num_tokens": 108071140.0,
      "step": 52070
    },
    {
      "entropy": 5.507170915603638,
      "epoch": 4.723784470246734,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029787522448816603,
      "loss": 5.0106,
      "mean_token_accuracy": 0.22013716101646424,
      "num_tokens": 108082599.0,
      "step": 52075
    },
    {
      "entropy": 5.565812492370606,
      "epoch": 4.7242380261248185,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029784303548936183,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22496211975812913,
      "num_tokens": 108092017.0,
      "step": 52080
    },
    {
      "entropy": 5.604690790176392,
      "epoch": 4.724691582002903,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002978108460181579,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21409368515014648,
      "num_tokens": 108103853.0,
      "step": 52085
    },
    {
      "entropy": 5.499201679229737,
      "epoch": 4.725145137880987,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002977786560752198,
      "loss": 4.9198,
      "mean_token_accuracy": 0.22652447521686553,
      "num_tokens": 108114727.0,
      "step": 52090
    },
    {
      "entropy": 5.476578187942505,
      "epoch": 4.725598693759071,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002977464656612133,
      "loss": 4.9007,
      "mean_token_accuracy": 0.22594528794288635,
      "num_tokens": 108125653.0,
      "step": 52095
    },
    {
      "entropy": 5.467841053009034,
      "epoch": 4.7260522496371555,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029771427477680415,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22339218556880952,
      "num_tokens": 108135281.0,
      "step": 52100
    },
    {
      "entropy": 5.51193413734436,
      "epoch": 4.72650580551524,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029768208342265796,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22418920546770096,
      "num_tokens": 108145005.0,
      "step": 52105
    },
    {
      "entropy": 5.523437213897705,
      "epoch": 4.726959361393323,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002976498915994406,
      "loss": 5.0288,
      "mean_token_accuracy": 0.21198291927576066,
      "num_tokens": 108156477.0,
      "step": 52110
    },
    {
      "entropy": 5.529740905761718,
      "epoch": 4.727412917271408,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029761769930781757,
      "loss": 4.9492,
      "mean_token_accuracy": 0.22652479261159897,
      "num_tokens": 108166947.0,
      "step": 52115
    },
    {
      "entropy": 5.467782640457154,
      "epoch": 4.727866473149492,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002975855065484548,
      "loss": 4.807,
      "mean_token_accuracy": 0.23662808239459993,
      "num_tokens": 108175889.0,
      "step": 52120
    },
    {
      "entropy": 5.540987539291382,
      "epoch": 4.728320029027576,
      "grad_norm": 1.03125,
      "learning_rate": 0.000297553313322018,
      "loss": 5.0012,
      "mean_token_accuracy": 0.21421878188848495,
      "num_tokens": 108186276.0,
      "step": 52125
    },
    {
      "entropy": 5.433714294433594,
      "epoch": 4.72877358490566,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029752111962917297,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2174909457564354,
      "num_tokens": 108196960.0,
      "step": 52130
    },
    {
      "entropy": 5.500071382522583,
      "epoch": 4.7292271407837445,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002974889254705854,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21918774247169495,
      "num_tokens": 108207442.0,
      "step": 52135
    },
    {
      "entropy": 5.509184074401856,
      "epoch": 4.729680696661829,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029745673084692116,
      "loss": 4.8824,
      "mean_token_accuracy": 0.23249851912260056,
      "num_tokens": 108217919.0,
      "step": 52140
    },
    {
      "entropy": 5.457738018035888,
      "epoch": 4.730134252539913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000297424535758846,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22616191059350968,
      "num_tokens": 108228106.0,
      "step": 52145
    },
    {
      "entropy": 5.466896343231201,
      "epoch": 4.730587808417997,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029739234020702564,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21878032237291337,
      "num_tokens": 108238398.0,
      "step": 52150
    },
    {
      "entropy": 5.524272346496582,
      "epoch": 4.731041364296082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029736014419212605,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21731851249933243,
      "num_tokens": 108248825.0,
      "step": 52155
    },
    {
      "entropy": 5.471959066390991,
      "epoch": 4.731494920174166,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029732794771481305,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22553227692842484,
      "num_tokens": 108258989.0,
      "step": 52160
    },
    {
      "entropy": 5.427928876876831,
      "epoch": 4.731948476052249,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002972957507757523,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2249559700489044,
      "num_tokens": 108268251.0,
      "step": 52165
    },
    {
      "entropy": 5.470182466506958,
      "epoch": 4.7324020319303335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029726355337560985,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21468491852283478,
      "num_tokens": 108278290.0,
      "step": 52170
    },
    {
      "entropy": 5.483646583557129,
      "epoch": 4.732855587808418,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029723135551505133,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21841174215078354,
      "num_tokens": 108288727.0,
      "step": 52175
    },
    {
      "entropy": 5.453590059280396,
      "epoch": 4.733309143686502,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029719915719474274,
      "loss": 4.9125,
      "mean_token_accuracy": 0.227685409784317,
      "num_tokens": 108299781.0,
      "step": 52180
    },
    {
      "entropy": 5.564180755615235,
      "epoch": 4.733762699564586,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002971669584153499,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21721042096614837,
      "num_tokens": 108310703.0,
      "step": 52185
    },
    {
      "entropy": 5.460237741470337,
      "epoch": 4.7342162554426706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002971347591775388,
      "loss": 4.884,
      "mean_token_accuracy": 0.22052497714757918,
      "num_tokens": 108321096.0,
      "step": 52190
    },
    {
      "entropy": 5.502922916412354,
      "epoch": 4.734669811320755,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002971025594819751,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2184264838695526,
      "num_tokens": 108331899.0,
      "step": 52195
    },
    {
      "entropy": 5.423063182830811,
      "epoch": 4.735123367198839,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029707035932932495,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2291702687740326,
      "num_tokens": 108342248.0,
      "step": 52200
    },
    {
      "entropy": 5.520014095306396,
      "epoch": 4.735576923076923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029703815872025413,
      "loss": 4.9848,
      "mean_token_accuracy": 0.21801053285598754,
      "num_tokens": 108355264.0,
      "step": 52205
    },
    {
      "entropy": 5.4837367057800295,
      "epoch": 4.736030478955008,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002970059576554285,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22193937003612518,
      "num_tokens": 108365929.0,
      "step": 52210
    },
    {
      "entropy": 5.516878843307495,
      "epoch": 4.736484034833092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002969737561355142,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2189727947115898,
      "num_tokens": 108376874.0,
      "step": 52215
    },
    {
      "entropy": 5.464402484893799,
      "epoch": 4.736937590711175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002969415541611769,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21912791877985,
      "num_tokens": 108387277.0,
      "step": 52220
    },
    {
      "entropy": 5.520644807815552,
      "epoch": 4.7373911465892595,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002969093517330827,
      "loss": 5.0395,
      "mean_token_accuracy": 0.21407714039087294,
      "num_tokens": 108399088.0,
      "step": 52225
    },
    {
      "entropy": 5.553883647918701,
      "epoch": 4.737844702467344,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002968771488518975,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21503780037164688,
      "num_tokens": 108409875.0,
      "step": 52230
    },
    {
      "entropy": 5.547718524932861,
      "epoch": 4.738298258345428,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002968449455182874,
      "loss": 4.9638,
      "mean_token_accuracy": 0.21943862438201905,
      "num_tokens": 108421133.0,
      "step": 52235
    },
    {
      "entropy": 5.6740494728088375,
      "epoch": 4.738751814223512,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029681274173291817,
      "loss": 5.1401,
      "mean_token_accuracy": 0.20525119453668594,
      "num_tokens": 108433408.0,
      "step": 52240
    },
    {
      "entropy": 5.473830461502075,
      "epoch": 4.739205370101597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029678053749645583,
      "loss": 4.854,
      "mean_token_accuracy": 0.22676481008529664,
      "num_tokens": 108443783.0,
      "step": 52245
    },
    {
      "entropy": 5.3422821998596195,
      "epoch": 4.739658925979681,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002967483328095665,
      "loss": 4.7942,
      "mean_token_accuracy": 0.229089917242527,
      "num_tokens": 108454770.0,
      "step": 52250
    },
    {
      "entropy": 5.455041933059692,
      "epoch": 4.740112481857765,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002967161276729161,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22599954456090926,
      "num_tokens": 108464812.0,
      "step": 52255
    },
    {
      "entropy": 5.514842987060547,
      "epoch": 4.740566037735849,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002966839220871707,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21652488112449647,
      "num_tokens": 108474642.0,
      "step": 52260
    },
    {
      "entropy": 5.431385231018067,
      "epoch": 4.741019593613933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002966517160529962,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22553370147943497,
      "num_tokens": 108484143.0,
      "step": 52265
    },
    {
      "entropy": 5.556461000442505,
      "epoch": 4.741473149492017,
      "grad_norm": 1.125,
      "learning_rate": 0.0002966195095710587,
      "loss": 5.0517,
      "mean_token_accuracy": 0.2114802837371826,
      "num_tokens": 108494850.0,
      "step": 52270
    },
    {
      "entropy": 5.512643051147461,
      "epoch": 4.741926705370101,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002965873026420242,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21448934823274612,
      "num_tokens": 108505441.0,
      "step": 52275
    },
    {
      "entropy": 5.527028656005859,
      "epoch": 4.742380261248186,
      "grad_norm": 1.125,
      "learning_rate": 0.00029655509526655885,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2121005728840828,
      "num_tokens": 108517350.0,
      "step": 52280
    },
    {
      "entropy": 5.451966714859009,
      "epoch": 4.74283381712627,
      "grad_norm": 1.125,
      "learning_rate": 0.00029652288744532856,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22282317280769348,
      "num_tokens": 108528472.0,
      "step": 52285
    },
    {
      "entropy": 5.504140710830688,
      "epoch": 4.743287373004354,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002964906791789995,
      "loss": 4.988,
      "mean_token_accuracy": 0.21872292459011078,
      "num_tokens": 108539834.0,
      "step": 52290
    },
    {
      "entropy": 5.578562927246094,
      "epoch": 4.743740928882438,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002964584704682377,
      "loss": 5.0245,
      "mean_token_accuracy": 0.22304878532886505,
      "num_tokens": 108551478.0,
      "step": 52295
    },
    {
      "entropy": 5.399234104156494,
      "epoch": 4.744194484760523,
      "grad_norm": 0.984375,
      "learning_rate": 0.00029642626131370936,
      "loss": 4.7612,
      "mean_token_accuracy": 0.23773835152387618,
      "num_tokens": 108560957.0,
      "step": 52300
    },
    {
      "entropy": 5.467911672592163,
      "epoch": 4.744648040638607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002963940517160804,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21856735199689864,
      "num_tokens": 108571146.0,
      "step": 52305
    },
    {
      "entropy": 5.498243141174316,
      "epoch": 4.745101596516691,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029636184167601695,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22440822422504425,
      "num_tokens": 108581085.0,
      "step": 52310
    },
    {
      "entropy": 5.4507440567016605,
      "epoch": 4.745555152394775,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029632963119418526,
      "loss": 4.8864,
      "mean_token_accuracy": 0.22616278976202012,
      "num_tokens": 108591103.0,
      "step": 52315
    },
    {
      "entropy": 5.491018295288086,
      "epoch": 4.746008708272859,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029629742027125134,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22114972919225692,
      "num_tokens": 108601433.0,
      "step": 52320
    },
    {
      "entropy": 5.465327501296997,
      "epoch": 4.746462264150943,
      "grad_norm": 1.125,
      "learning_rate": 0.00029626520890788127,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22479272484779358,
      "num_tokens": 108611728.0,
      "step": 52325
    },
    {
      "entropy": 5.480198955535888,
      "epoch": 4.746915820029027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029623299710474134,
      "loss": 4.9115,
      "mean_token_accuracy": 0.22393460422754288,
      "num_tokens": 108621521.0,
      "step": 52330
    },
    {
      "entropy": 5.485647869110108,
      "epoch": 4.747369375907112,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029620078486249763,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2176494374871254,
      "num_tokens": 108632921.0,
      "step": 52335
    },
    {
      "entropy": 5.5155998229980465,
      "epoch": 4.747822931785196,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002961685721818163,
      "loss": 4.9361,
      "mean_token_accuracy": 0.22355274707078934,
      "num_tokens": 108643244.0,
      "step": 52340
    },
    {
      "entropy": 5.506576490402222,
      "epoch": 4.74827648766328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002961363590633634,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22796093672513962,
      "num_tokens": 108653036.0,
      "step": 52345
    },
    {
      "entropy": 5.535421323776245,
      "epoch": 4.748730043541364,
      "grad_norm": 1.0,
      "learning_rate": 0.0002961041455078053,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2267568439245224,
      "num_tokens": 108663317.0,
      "step": 52350
    },
    {
      "entropy": 5.498347473144531,
      "epoch": 4.749183599419449,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002960719315158081,
      "loss": 4.865,
      "mean_token_accuracy": 0.22356354147195817,
      "num_tokens": 108673657.0,
      "step": 52355
    },
    {
      "entropy": 5.464849424362183,
      "epoch": 4.749637155297533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002960397170880378,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2163007453083992,
      "num_tokens": 108683831.0,
      "step": 52360
    },
    {
      "entropy": 5.45690689086914,
      "epoch": 4.750090711175617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029600750222516097,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2187454342842102,
      "num_tokens": 108694062.0,
      "step": 52365
    },
    {
      "entropy": 5.457013463973999,
      "epoch": 4.7505442670537015,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00029597528692784365,
      "loss": 4.904,
      "mean_token_accuracy": 0.223398120701313,
      "num_tokens": 108704712.0,
      "step": 52370
    },
    {
      "entropy": 5.478539228439331,
      "epoch": 4.750997822931785,
      "grad_norm": 1.0,
      "learning_rate": 0.00029594307119675195,
      "loss": 4.889,
      "mean_token_accuracy": 0.22490335255861282,
      "num_tokens": 108714741.0,
      "step": 52375
    },
    {
      "entropy": 5.531120872497558,
      "epoch": 4.751451378809869,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002959108550325523,
      "loss": 4.9901,
      "mean_token_accuracy": 0.22630801647901536,
      "num_tokens": 108724860.0,
      "step": 52380
    },
    {
      "entropy": 5.431674003601074,
      "epoch": 4.751904934687953,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029587863843591076,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23203105032444,
      "num_tokens": 108735527.0,
      "step": 52385
    },
    {
      "entropy": 5.459289264678955,
      "epoch": 4.752358490566038,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029584642140749375,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20949258655309677,
      "num_tokens": 108745748.0,
      "step": 52390
    },
    {
      "entropy": 5.532995891571045,
      "epoch": 4.752812046444122,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00029581420394796733,
      "loss": 4.9405,
      "mean_token_accuracy": 0.223834764957428,
      "num_tokens": 108756635.0,
      "step": 52395
    },
    {
      "entropy": 5.506797742843628,
      "epoch": 4.753265602322206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029578198605799796,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22417925745248796,
      "num_tokens": 108765912.0,
      "step": 52400
    },
    {
      "entropy": 5.518583869934082,
      "epoch": 4.7537191582002905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029574976773825176,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2274146482348442,
      "num_tokens": 108775831.0,
      "step": 52405
    },
    {
      "entropy": 5.4731968402862545,
      "epoch": 4.754172714078375,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000295717548989395,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2241987928748131,
      "num_tokens": 108785937.0,
      "step": 52410
    },
    {
      "entropy": 5.446841526031494,
      "epoch": 4.754626269956459,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002956853298120942,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22704738080501558,
      "num_tokens": 108796289.0,
      "step": 52415
    },
    {
      "entropy": 5.515114688873291,
      "epoch": 4.755079825834542,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029565311020701544,
      "loss": 4.964,
      "mean_token_accuracy": 0.21557584404945374,
      "num_tokens": 108806040.0,
      "step": 52420
    },
    {
      "entropy": 5.5106954097747805,
      "epoch": 4.755533381712627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029562089017482515,
      "loss": 4.953,
      "mean_token_accuracy": 0.2192101448774338,
      "num_tokens": 108815440.0,
      "step": 52425
    },
    {
      "entropy": 5.556234073638916,
      "epoch": 4.755986937590711,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002955886697161895,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21606413275003433,
      "num_tokens": 108826059.0,
      "step": 52430
    },
    {
      "entropy": 5.531109142303467,
      "epoch": 4.756440493468795,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029555644883177496,
      "loss": 4.903,
      "mean_token_accuracy": 0.22589529752731324,
      "num_tokens": 108836373.0,
      "step": 52435
    },
    {
      "entropy": 5.44771523475647,
      "epoch": 4.756894049346879,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029552422752224787,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23270179629325866,
      "num_tokens": 108847364.0,
      "step": 52440
    },
    {
      "entropy": 5.42327733039856,
      "epoch": 4.757347605224964,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029549200578827444,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21961222141981124,
      "num_tokens": 108856893.0,
      "step": 52445
    },
    {
      "entropy": 5.5134154796600345,
      "epoch": 4.757801161103048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002954597836305212,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22851149439811708,
      "num_tokens": 108867218.0,
      "step": 52450
    },
    {
      "entropy": 5.4865710735321045,
      "epoch": 4.758254716981132,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029542756104965434,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2209153100848198,
      "num_tokens": 108876938.0,
      "step": 52455
    },
    {
      "entropy": 5.506293392181396,
      "epoch": 4.7587082728592165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029539533804634027,
      "loss": 4.9201,
      "mean_token_accuracy": 0.2212717205286026,
      "num_tokens": 108887831.0,
      "step": 52460
    },
    {
      "entropy": 5.437268257141113,
      "epoch": 4.759161828737301,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029536311462124553,
      "loss": 4.883,
      "mean_token_accuracy": 0.23057824224233628,
      "num_tokens": 108897634.0,
      "step": 52465
    },
    {
      "entropy": 5.457991743087769,
      "epoch": 4.759615384615385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002953308907750363,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22301412969827653,
      "num_tokens": 108907976.0,
      "step": 52470
    },
    {
      "entropy": 5.483482885360718,
      "epoch": 4.760068940493468,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002952986665083792,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2212643399834633,
      "num_tokens": 108918755.0,
      "step": 52475
    },
    {
      "entropy": 5.493503999710083,
      "epoch": 4.760522496371553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002952664418219404,
      "loss": 4.9227,
      "mean_token_accuracy": 0.22590409964323044,
      "num_tokens": 108928349.0,
      "step": 52480
    },
    {
      "entropy": 5.501414203643799,
      "epoch": 4.760976052249637,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00029523421671638644,
      "loss": 4.977,
      "mean_token_accuracy": 0.21380869746208192,
      "num_tokens": 108938429.0,
      "step": 52485
    },
    {
      "entropy": 5.524934482574463,
      "epoch": 4.761429608127721,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002952019911923837,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21943130046129228,
      "num_tokens": 108948378.0,
      "step": 52490
    },
    {
      "entropy": 5.452820539474487,
      "epoch": 4.7618831640058055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029516976525059864,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22503216862678527,
      "num_tokens": 108958667.0,
      "step": 52495
    },
    {
      "entropy": 5.557079315185547,
      "epoch": 4.76233671988389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002951375388916977,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2175732284784317,
      "num_tokens": 108968955.0,
      "step": 52500
    },
    {
      "entropy": 5.48816556930542,
      "epoch": 4.762790275761974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002951053121163474,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2205175682902336,
      "num_tokens": 108979565.0,
      "step": 52505
    },
    {
      "entropy": 5.4887782573699955,
      "epoch": 4.763243831640058,
      "grad_norm": 1.078125,
      "learning_rate": 0.000295073084925214,
      "loss": 4.949,
      "mean_token_accuracy": 0.21613194644451142,
      "num_tokens": 108989547.0,
      "step": 52510
    },
    {
      "entropy": 5.446226453781128,
      "epoch": 4.7636973875181425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029504085731896413,
      "loss": 4.903,
      "mean_token_accuracy": 0.21974362432956696,
      "num_tokens": 109000286.0,
      "step": 52515
    },
    {
      "entropy": 5.5266036033630375,
      "epoch": 4.764150943396227,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002950086292982642,
      "loss": 4.8738,
      "mean_token_accuracy": 0.227171827852726,
      "num_tokens": 109010676.0,
      "step": 52520
    },
    {
      "entropy": 5.570739126205444,
      "epoch": 4.764604499274311,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029497640086378077,
      "loss": 5.0069,
      "mean_token_accuracy": 0.21162082701921464,
      "num_tokens": 109020800.0,
      "step": 52525
    },
    {
      "entropy": 5.509361982345581,
      "epoch": 4.7650580551523944,
      "grad_norm": 1.0,
      "learning_rate": 0.00029494417201618027,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2185629665851593,
      "num_tokens": 109031158.0,
      "step": 52530
    },
    {
      "entropy": 5.493510770797729,
      "epoch": 4.765511611030479,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002949119427561292,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22346618473529817,
      "num_tokens": 109041664.0,
      "step": 52535
    },
    {
      "entropy": 5.500121784210205,
      "epoch": 4.765965166908563,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029487971308429415,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2287719026207924,
      "num_tokens": 109052633.0,
      "step": 52540
    },
    {
      "entropy": 5.463460159301758,
      "epoch": 4.766418722786647,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002948474830013414,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22805143296718597,
      "num_tokens": 109062687.0,
      "step": 52545
    },
    {
      "entropy": 5.558494520187378,
      "epoch": 4.7668722786647315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029481525250793777,
      "loss": 4.9844,
      "mean_token_accuracy": 0.22054018676280976,
      "num_tokens": 109073348.0,
      "step": 52550
    },
    {
      "entropy": 5.471759891510009,
      "epoch": 4.767325834542816,
      "grad_norm": 1.125,
      "learning_rate": 0.0002947830216047497,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21687054485082627,
      "num_tokens": 109083276.0,
      "step": 52555
    },
    {
      "entropy": 5.447759389877319,
      "epoch": 4.7677793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029475079029244356,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22003519237041474,
      "num_tokens": 109094342.0,
      "step": 52560
    },
    {
      "entropy": 5.5125412940979,
      "epoch": 4.768232946298984,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00029471855857168616,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2255170986056328,
      "num_tokens": 109105178.0,
      "step": 52565
    },
    {
      "entropy": 5.56022367477417,
      "epoch": 4.768686502177069,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002946863264431439,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21164515912532805,
      "num_tokens": 109116205.0,
      "step": 52570
    },
    {
      "entropy": 5.478892135620117,
      "epoch": 4.769140058055152,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002946540939074834,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21910533010959626,
      "num_tokens": 109126299.0,
      "step": 52575
    },
    {
      "entropy": 5.534829568862915,
      "epoch": 4.769593613933236,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00029462186096537116,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21910370141267776,
      "num_tokens": 109137675.0,
      "step": 52580
    },
    {
      "entropy": 5.529242849349975,
      "epoch": 4.7700471698113205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029458962761747394,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2256185919046402,
      "num_tokens": 109147716.0,
      "step": 52585
    },
    {
      "entropy": 5.528822898864746,
      "epoch": 4.770500725689405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002945573938644582,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21762220114469527,
      "num_tokens": 109159211.0,
      "step": 52590
    },
    {
      "entropy": 5.479215478897094,
      "epoch": 4.770954281567489,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029452515970699054,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21710111498832702,
      "num_tokens": 109169490.0,
      "step": 52595
    },
    {
      "entropy": 5.527347993850708,
      "epoch": 4.771407837445573,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00029449292514573767,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21463383585214615,
      "num_tokens": 109179687.0,
      "step": 52600
    },
    {
      "entropy": 5.491201543807984,
      "epoch": 4.7718613933236576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029446069018136613,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22277799248695374,
      "num_tokens": 109189415.0,
      "step": 52605
    },
    {
      "entropy": 5.442613840103149,
      "epoch": 4.772314949201742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002944284548145424,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22883726209402083,
      "num_tokens": 109199281.0,
      "step": 52610
    },
    {
      "entropy": 5.466050910949707,
      "epoch": 4.772768505079826,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002943962190459335,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22881308645009996,
      "num_tokens": 109208347.0,
      "step": 52615
    },
    {
      "entropy": 5.472415018081665,
      "epoch": 4.77322206095791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029436398287620573,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21719896346330642,
      "num_tokens": 109218026.0,
      "step": 52620
    },
    {
      "entropy": 5.44177360534668,
      "epoch": 4.773675616835995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002943317463060259,
      "loss": 4.8949,
      "mean_token_accuracy": 0.21816908270120622,
      "num_tokens": 109229457.0,
      "step": 52625
    },
    {
      "entropy": 5.4849803924560545,
      "epoch": 4.774129172714078,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002942995093360606,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22519377619028091,
      "num_tokens": 109239504.0,
      "step": 52630
    },
    {
      "entropy": 5.384753942489624,
      "epoch": 4.774582728592162,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002942672719669765,
      "loss": 4.8031,
      "mean_token_accuracy": 0.2309288814663887,
      "num_tokens": 109249745.0,
      "step": 52635
    },
    {
      "entropy": 5.4371696472167965,
      "epoch": 4.7750362844702465,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002942350341994404,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22550443112850188,
      "num_tokens": 109259767.0,
      "step": 52640
    },
    {
      "entropy": 5.620873832702637,
      "epoch": 4.775489840348331,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029420279603411875,
      "loss": 5.1028,
      "mean_token_accuracy": 0.2097809836268425,
      "num_tokens": 109269611.0,
      "step": 52645
    },
    {
      "entropy": 5.627543020248413,
      "epoch": 4.775943396226415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002941705574716785,
      "loss": 5.0139,
      "mean_token_accuracy": 0.21355631500482558,
      "num_tokens": 109281396.0,
      "step": 52650
    },
    {
      "entropy": 5.531946897506714,
      "epoch": 4.776396952104499,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002941383185127862,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2206423655152321,
      "num_tokens": 109292346.0,
      "step": 52655
    },
    {
      "entropy": 5.567822027206421,
      "epoch": 4.776850507982584,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029410607915810854,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21609701812267304,
      "num_tokens": 109302966.0,
      "step": 52660
    },
    {
      "entropy": 5.40581521987915,
      "epoch": 4.777304063860668,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002940738394083124,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22364492267370223,
      "num_tokens": 109312558.0,
      "step": 52665
    },
    {
      "entropy": 5.527387428283691,
      "epoch": 4.777757619738752,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029404159926406436,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2193790480494499,
      "num_tokens": 109322364.0,
      "step": 52670
    },
    {
      "entropy": 5.575751543045044,
      "epoch": 4.7782111756168355,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002940093587260312,
      "loss": 5.0073,
      "mean_token_accuracy": 0.21996435821056365,
      "num_tokens": 109332449.0,
      "step": 52675
    },
    {
      "entropy": 5.536706161499024,
      "epoch": 4.778664731494921,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029397711779487954,
      "loss": 4.9192,
      "mean_token_accuracy": 0.224237360060215,
      "num_tokens": 109343262.0,
      "step": 52680
    },
    {
      "entropy": 5.449114179611206,
      "epoch": 4.779118287373004,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029394487647127645,
      "loss": 4.8566,
      "mean_token_accuracy": 0.2278436928987503,
      "num_tokens": 109352951.0,
      "step": 52685
    },
    {
      "entropy": 5.420614385604859,
      "epoch": 4.779571843251088,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029391263475588844,
      "loss": 5.0109,
      "mean_token_accuracy": 0.21873986721038818,
      "num_tokens": 109363913.0,
      "step": 52690
    },
    {
      "entropy": 5.456114053726196,
      "epoch": 4.780025399129173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029388039264938224,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22255000174045564,
      "num_tokens": 109374405.0,
      "step": 52695
    },
    {
      "entropy": 5.574003791809082,
      "epoch": 4.780478955007257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002938481501524247,
      "loss": 5.0046,
      "mean_token_accuracy": 0.21413532495498658,
      "num_tokens": 109385696.0,
      "step": 52700
    },
    {
      "entropy": 5.530571317672729,
      "epoch": 4.780932510885341,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029381590726568267,
      "loss": 4.9617,
      "mean_token_accuracy": 0.22094358652830123,
      "num_tokens": 109395061.0,
      "step": 52705
    },
    {
      "entropy": 5.495776033401489,
      "epoch": 4.781386066763425,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00029378366398982286,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21628472954034805,
      "num_tokens": 109406152.0,
      "step": 52710
    },
    {
      "entropy": 5.563758611679077,
      "epoch": 4.78183962264151,
      "grad_norm": 1.0,
      "learning_rate": 0.00029375142032551216,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2144368976354599,
      "num_tokens": 109416596.0,
      "step": 52715
    },
    {
      "entropy": 5.585849285125732,
      "epoch": 4.782293178519594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00029371917627341726,
      "loss": 4.9998,
      "mean_token_accuracy": 0.21506147235631942,
      "num_tokens": 109426169.0,
      "step": 52720
    },
    {
      "entropy": 5.549559307098389,
      "epoch": 4.782746734397678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000293686931834205,
      "loss": 4.9603,
      "mean_token_accuracy": 0.22521220594644548,
      "num_tokens": 109436359.0,
      "step": 52725
    },
    {
      "entropy": 5.469870948791504,
      "epoch": 4.7832002902757615,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002936546870085422,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21868258118629455,
      "num_tokens": 109446050.0,
      "step": 52730
    },
    {
      "entropy": 5.5514427661895756,
      "epoch": 4.783653846153846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029362244179709576,
      "loss": 4.9784,
      "mean_token_accuracy": 0.22177448123693466,
      "num_tokens": 109457124.0,
      "step": 52735
    },
    {
      "entropy": 5.443732738494873,
      "epoch": 4.78410740203193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029359019620053253,
      "loss": 4.8403,
      "mean_token_accuracy": 0.2237445145845413,
      "num_tokens": 109467642.0,
      "step": 52740
    },
    {
      "entropy": 5.5143656730651855,
      "epoch": 4.784560957910014,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002935579502195192,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21531228572130204,
      "num_tokens": 109477877.0,
      "step": 52745
    },
    {
      "entropy": 5.492346668243409,
      "epoch": 4.785014513788099,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00029352570385472276,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22153888940811156,
      "num_tokens": 109489302.0,
      "step": 52750
    },
    {
      "entropy": 5.4734595775604244,
      "epoch": 4.785468069666183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002934934571068101,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22341156899929046,
      "num_tokens": 109500023.0,
      "step": 52755
    },
    {
      "entropy": 5.442930078506469,
      "epoch": 4.785921625544267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000293461209976448,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22326204776763917,
      "num_tokens": 109510588.0,
      "step": 52760
    },
    {
      "entropy": 5.561079311370849,
      "epoch": 4.786375181422351,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002934289624643034,
      "loss": 4.969,
      "mean_token_accuracy": 0.21650516986846924,
      "num_tokens": 109520162.0,
      "step": 52765
    },
    {
      "entropy": 5.522705316543579,
      "epoch": 4.786828737300436,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002933967145710431,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2215215355157852,
      "num_tokens": 109530155.0,
      "step": 52770
    },
    {
      "entropy": 5.497471570968628,
      "epoch": 4.78728229317852,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002933644662973341,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2252976968884468,
      "num_tokens": 109539897.0,
      "step": 52775
    },
    {
      "entropy": 5.4756018161773685,
      "epoch": 4.787735849056604,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029333221764384326,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22061517089605331,
      "num_tokens": 109549995.0,
      "step": 52780
    },
    {
      "entropy": 5.504759883880615,
      "epoch": 4.788189404934688,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029329996861123745,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21992766559123994,
      "num_tokens": 109560446.0,
      "step": 52785
    },
    {
      "entropy": 5.44918646812439,
      "epoch": 4.788642960812772,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00029326771920018367,
      "loss": 4.882,
      "mean_token_accuracy": 0.2233288049697876,
      "num_tokens": 109570864.0,
      "step": 52790
    },
    {
      "entropy": 5.586102104187011,
      "epoch": 4.789096516690856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029323546941134875,
      "loss": 4.9559,
      "mean_token_accuracy": 0.22462151795625687,
      "num_tokens": 109581015.0,
      "step": 52795
    },
    {
      "entropy": 5.480606937408448,
      "epoch": 4.78955007256894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029320321924539974,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21978964060544967,
      "num_tokens": 109591695.0,
      "step": 52800
    },
    {
      "entropy": 5.447766208648682,
      "epoch": 4.790003628447025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002931709687030035,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22672177702188492,
      "num_tokens": 109602274.0,
      "step": 52805
    },
    {
      "entropy": 5.557154273986816,
      "epoch": 4.790457184325109,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029313871778482693,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2248819187283516,
      "num_tokens": 109612995.0,
      "step": 52810
    },
    {
      "entropy": 5.499710893630981,
      "epoch": 4.790910740203193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002931064664915371,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22039807885885238,
      "num_tokens": 109623133.0,
      "step": 52815
    },
    {
      "entropy": 5.408206129074097,
      "epoch": 4.7913642960812775,
      "grad_norm": 1.046875,
      "learning_rate": 0.000293074214823801,
      "loss": 4.873,
      "mean_token_accuracy": 0.2262205258011818,
      "num_tokens": 109634312.0,
      "step": 52820
    },
    {
      "entropy": 5.513941764831543,
      "epoch": 4.791817851959362,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029304196278228544,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20868304073810579,
      "num_tokens": 109644333.0,
      "step": 52825
    },
    {
      "entropy": 5.477426767349243,
      "epoch": 4.792271407837445,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002930097103676575,
      "loss": 4.8455,
      "mean_token_accuracy": 0.22492458969354628,
      "num_tokens": 109654730.0,
      "step": 52830
    },
    {
      "entropy": 5.461199235916138,
      "epoch": 4.79272496371553,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029297745758058414,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2240580901503563,
      "num_tokens": 109664867.0,
      "step": 52835
    },
    {
      "entropy": 5.521033382415771,
      "epoch": 4.793178519593614,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029294520442173244,
      "loss": 4.978,
      "mean_token_accuracy": 0.22023186534643174,
      "num_tokens": 109675090.0,
      "step": 52840
    },
    {
      "entropy": 5.445932102203369,
      "epoch": 4.793632075471698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002929129508917693,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2320157378911972,
      "num_tokens": 109685046.0,
      "step": 52845
    },
    {
      "entropy": 5.588843297958374,
      "epoch": 4.794085631349782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002928806969913618,
      "loss": 4.9753,
      "mean_token_accuracy": 0.21928717643022538,
      "num_tokens": 109696240.0,
      "step": 52850
    },
    {
      "entropy": 5.49670376777649,
      "epoch": 4.794539187227866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002928484427211769,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21479919403791428,
      "num_tokens": 109707676.0,
      "step": 52855
    },
    {
      "entropy": 5.561608076095581,
      "epoch": 4.794992743105951,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002928161880818817,
      "loss": 5.0161,
      "mean_token_accuracy": 0.21642168164253234,
      "num_tokens": 109718168.0,
      "step": 52860
    },
    {
      "entropy": 5.494682264328003,
      "epoch": 4.795446298984035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002927839330741431,
      "loss": 4.8824,
      "mean_token_accuracy": 0.22129243910312651,
      "num_tokens": 109728624.0,
      "step": 52865
    },
    {
      "entropy": 5.595055675506591,
      "epoch": 4.795899854862119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029275167769862834,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21029520332813262,
      "num_tokens": 109738301.0,
      "step": 52870
    },
    {
      "entropy": 5.5356913089752195,
      "epoch": 4.7963534107402035,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002927194219560043,
      "loss": 5.002,
      "mean_token_accuracy": 0.2135516494512558,
      "num_tokens": 109749269.0,
      "step": 52875
    },
    {
      "entropy": 5.476754188537598,
      "epoch": 4.796806966618288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029268716584693804,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22285651415586472,
      "num_tokens": 109758335.0,
      "step": 52880
    },
    {
      "entropy": 5.486256742477417,
      "epoch": 4.797260522496371,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002926549093720968,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22424013018608094,
      "num_tokens": 109768848.0,
      "step": 52885
    },
    {
      "entropy": 5.454851150512695,
      "epoch": 4.797714078374455,
      "grad_norm": 0.984375,
      "learning_rate": 0.00029262265253214754,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21794513314962388,
      "num_tokens": 109779331.0,
      "step": 52890
    },
    {
      "entropy": 5.52949595451355,
      "epoch": 4.79816763425254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029259039532775725,
      "loss": 4.9181,
      "mean_token_accuracy": 0.22026457339525224,
      "num_tokens": 109789331.0,
      "step": 52895
    },
    {
      "entropy": 5.538748359680175,
      "epoch": 4.798621190130624,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002925581377595931,
      "loss": 4.8818,
      "mean_token_accuracy": 0.2268442451953888,
      "num_tokens": 109798630.0,
      "step": 52900
    },
    {
      "entropy": 5.44479022026062,
      "epoch": 4.799074746008708,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029252587982832225,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22784989029169084,
      "num_tokens": 109808501.0,
      "step": 52905
    },
    {
      "entropy": 5.467966985702515,
      "epoch": 4.7995283018867925,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002924936215346117,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22745104432106017,
      "num_tokens": 109818847.0,
      "step": 52910
    },
    {
      "entropy": 5.54151554107666,
      "epoch": 4.799981857764877,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029246136287912865,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2198122814297676,
      "num_tokens": 109830017.0,
      "step": 52915
    },
    {
      "entropy": 5.492886972427368,
      "epoch": 4.800435413642961,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002924291038625402,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21905387043952942,
      "num_tokens": 109839795.0,
      "step": 52920
    },
    {
      "entropy": 5.49671745300293,
      "epoch": 4.800888969521045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029239684448551336,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22454091310501098,
      "num_tokens": 109850202.0,
      "step": 52925
    },
    {
      "entropy": 5.435492753982544,
      "epoch": 4.8013425253991295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002923645847487153,
      "loss": 4.9365,
      "mean_token_accuracy": 0.22263592779636382,
      "num_tokens": 109860625.0,
      "step": 52930
    },
    {
      "entropy": 5.501131391525268,
      "epoch": 4.801796081277214,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002923323246528133,
      "loss": 4.908,
      "mean_token_accuracy": 0.22579710632562638,
      "num_tokens": 109871241.0,
      "step": 52935
    },
    {
      "entropy": 5.5447601795196535,
      "epoch": 4.802249637155297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029230006419847434,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21876299530267715,
      "num_tokens": 109881627.0,
      "step": 52940
    },
    {
      "entropy": 5.511211395263672,
      "epoch": 4.8027031930333814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002922678033863656,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21997952163219453,
      "num_tokens": 109893247.0,
      "step": 52945
    },
    {
      "entropy": 5.484923028945923,
      "epoch": 4.803156748911466,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002922355422171544,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21889303475618363,
      "num_tokens": 109904112.0,
      "step": 52950
    },
    {
      "entropy": 5.451795768737793,
      "epoch": 4.80361030478955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029220328069150774,
      "loss": 4.8084,
      "mean_token_accuracy": 0.23551703989505768,
      "num_tokens": 109913854.0,
      "step": 52955
    },
    {
      "entropy": 5.568061876296997,
      "epoch": 4.804063860667634,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029217101881009284,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2153307467699051,
      "num_tokens": 109923947.0,
      "step": 52960
    },
    {
      "entropy": 5.496352434158325,
      "epoch": 4.8045174165457185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002921387565735769,
      "loss": 4.9188,
      "mean_token_accuracy": 0.2221384510397911,
      "num_tokens": 109934590.0,
      "step": 52965
    },
    {
      "entropy": 5.462561845779419,
      "epoch": 4.804970972423803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002921064939826272,
      "loss": 4.939,
      "mean_token_accuracy": 0.2211701065301895,
      "num_tokens": 109944224.0,
      "step": 52970
    },
    {
      "entropy": 5.494122505187988,
      "epoch": 4.805424528301887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002920742310379108,
      "loss": 4.9674,
      "mean_token_accuracy": 0.22000376284122466,
      "num_tokens": 109954414.0,
      "step": 52975
    },
    {
      "entropy": 5.556624984741211,
      "epoch": 4.805878084179971,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029204196774009476,
      "loss": 4.9491,
      "mean_token_accuracy": 0.2236790731549263,
      "num_tokens": 109964304.0,
      "step": 52980
    },
    {
      "entropy": 5.574713230133057,
      "epoch": 4.806331640058055,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00029200970408984673,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20797566920518876,
      "num_tokens": 109975401.0,
      "step": 52985
    },
    {
      "entropy": 5.566577768325805,
      "epoch": 4.806785195936139,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002919774400878336,
      "loss": 4.9896,
      "mean_token_accuracy": 0.2222548693418503,
      "num_tokens": 109985964.0,
      "step": 52990
    },
    {
      "entropy": 5.564922571182251,
      "epoch": 4.807238751814223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002919451757347226,
      "loss": 4.9472,
      "mean_token_accuracy": 0.22143390029668808,
      "num_tokens": 109995476.0,
      "step": 52995
    },
    {
      "entropy": 5.522109842300415,
      "epoch": 4.8076923076923075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002919129110311812,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21973171085119247,
      "num_tokens": 110005238.0,
      "step": 53000
    },
    {
      "entropy": 5.530682373046875,
      "epoch": 4.808145863570392,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002918806459778764,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20393204391002656,
      "num_tokens": 110016120.0,
      "step": 53005
    },
    {
      "entropy": 5.517846632003784,
      "epoch": 4.808599419448476,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002918483805754756,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21533962786197663,
      "num_tokens": 110026739.0,
      "step": 53010
    },
    {
      "entropy": 5.574581050872803,
      "epoch": 4.80905297532656,
      "grad_norm": 1.125,
      "learning_rate": 0.0002918161148246459,
      "loss": 5.0456,
      "mean_token_accuracy": 0.2173927903175354,
      "num_tokens": 110037491.0,
      "step": 53015
    },
    {
      "entropy": 5.524920177459717,
      "epoch": 4.8095065312046446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029178384872605465,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22003861665725707,
      "num_tokens": 110048550.0,
      "step": 53020
    },
    {
      "entropy": 5.497657251358032,
      "epoch": 4.809960087082729,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029175158228036924,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21888701617717743,
      "num_tokens": 110058472.0,
      "step": 53025
    },
    {
      "entropy": 5.468406820297242,
      "epoch": 4.810413642960813,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002917193154882568,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22800790816545485,
      "num_tokens": 110070766.0,
      "step": 53030
    },
    {
      "entropy": 5.545467948913574,
      "epoch": 4.810867198838897,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002916870483503847,
      "loss": 4.9572,
      "mean_token_accuracy": 0.2238215908408165,
      "num_tokens": 110081739.0,
      "step": 53035
    },
    {
      "entropy": 5.496158647537231,
      "epoch": 4.811320754716981,
      "grad_norm": 1.0,
      "learning_rate": 0.00029165478086742006,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21987418979406356,
      "num_tokens": 110091075.0,
      "step": 53040
    },
    {
      "entropy": 5.455555009841919,
      "epoch": 4.811774310595065,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029162251304003044,
      "loss": 4.8895,
      "mean_token_accuracy": 0.2274573802947998,
      "num_tokens": 110100530.0,
      "step": 53045
    },
    {
      "entropy": 5.458979845046997,
      "epoch": 4.812227866473149,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002915902448688829,
      "loss": 4.871,
      "mean_token_accuracy": 0.22266223579645156,
      "num_tokens": 110110866.0,
      "step": 53050
    },
    {
      "entropy": 5.498134517669678,
      "epoch": 4.8126814223512335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029155797635464493,
      "loss": 4.9623,
      "mean_token_accuracy": 0.22287888526916505,
      "num_tokens": 110121344.0,
      "step": 53055
    },
    {
      "entropy": 5.441980266571045,
      "epoch": 4.813134978229318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002915257074979838,
      "loss": 4.834,
      "mean_token_accuracy": 0.22484128326177596,
      "num_tokens": 110130529.0,
      "step": 53060
    },
    {
      "entropy": 5.517879962921143,
      "epoch": 4.813588534107402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002914934382995668,
      "loss": 4.8693,
      "mean_token_accuracy": 0.223667711019516,
      "num_tokens": 110141536.0,
      "step": 53065
    },
    {
      "entropy": 5.539094591140747,
      "epoch": 4.814042089985486,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002914611687600613,
      "loss": 4.9766,
      "mean_token_accuracy": 0.22606740295886993,
      "num_tokens": 110152364.0,
      "step": 53070
    },
    {
      "entropy": 5.467326641082764,
      "epoch": 4.814495645863571,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002914288988801347,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22458479255437852,
      "num_tokens": 110163800.0,
      "step": 53075
    },
    {
      "entropy": 5.426185655593872,
      "epoch": 4.814949201741655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002913966286604541,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22011211812496184,
      "num_tokens": 110174590.0,
      "step": 53080
    },
    {
      "entropy": 5.55316014289856,
      "epoch": 4.815402757619739,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029136435810168726,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21838787645101548,
      "num_tokens": 110185142.0,
      "step": 53085
    },
    {
      "entropy": 5.433688163757324,
      "epoch": 4.815856313497823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002913320872045012,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23755647391080856,
      "num_tokens": 110194013.0,
      "step": 53090
    },
    {
      "entropy": 5.468377256393433,
      "epoch": 4.816309869375907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029129981596956353,
      "loss": 4.8908,
      "mean_token_accuracy": 0.23432908952236176,
      "num_tokens": 110204087.0,
      "step": 53095
    },
    {
      "entropy": 5.523484086990356,
      "epoch": 4.816763425253991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029126754439754137,
      "loss": 4.9372,
      "mean_token_accuracy": 0.2210382491350174,
      "num_tokens": 110214066.0,
      "step": 53100
    },
    {
      "entropy": 5.504038763046265,
      "epoch": 4.817216981132075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002912352724891023,
      "loss": 4.9147,
      "mean_token_accuracy": 0.230015829205513,
      "num_tokens": 110224378.0,
      "step": 53105
    },
    {
      "entropy": 5.452801036834717,
      "epoch": 4.81767053701016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002912030002449137,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22173485308885574,
      "num_tokens": 110235493.0,
      "step": 53110
    },
    {
      "entropy": 5.455511808395386,
      "epoch": 4.818124092888244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029117072766564285,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21360934376716614,
      "num_tokens": 110246953.0,
      "step": 53115
    },
    {
      "entropy": 5.464690780639648,
      "epoch": 4.818577648766328,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002911384547519573,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23481296747922897,
      "num_tokens": 110257387.0,
      "step": 53120
    },
    {
      "entropy": 5.480072736740112,
      "epoch": 4.819031204644412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029110618150452437,
      "loss": 4.93,
      "mean_token_accuracy": 0.2265430673956871,
      "num_tokens": 110268152.0,
      "step": 53125
    },
    {
      "entropy": 5.4890368461608885,
      "epoch": 4.819484760522497,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029107390792401145,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22170277833938598,
      "num_tokens": 110277933.0,
      "step": 53130
    },
    {
      "entropy": 5.471178913116455,
      "epoch": 4.819938316400581,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002910416340110861,
      "loss": 4.8198,
      "mean_token_accuracy": 0.22850697338581086,
      "num_tokens": 110288219.0,
      "step": 53135
    },
    {
      "entropy": 5.506328582763672,
      "epoch": 4.820391872278664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002910093597664156,
      "loss": 4.9392,
      "mean_token_accuracy": 0.22149244248867034,
      "num_tokens": 110299209.0,
      "step": 53140
    },
    {
      "entropy": 5.495154285430909,
      "epoch": 4.8208454281567485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002909770851906675,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22347409129142762,
      "num_tokens": 110309482.0,
      "step": 53145
    },
    {
      "entropy": 5.479849100112915,
      "epoch": 4.821298984034833,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029094481028450926,
      "loss": 4.88,
      "mean_token_accuracy": 0.22872833609580995,
      "num_tokens": 110319499.0,
      "step": 53150
    },
    {
      "entropy": 5.467418766021728,
      "epoch": 4.821752539912917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002909125350486082,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22905947566032409,
      "num_tokens": 110329453.0,
      "step": 53155
    },
    {
      "entropy": 5.46168155670166,
      "epoch": 4.822206095791001,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002908802594836319,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21841447353363036,
      "num_tokens": 110339343.0,
      "step": 53160
    },
    {
      "entropy": 5.514559268951416,
      "epoch": 4.822659651669086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002908479835902477,
      "loss": 4.9861,
      "mean_token_accuracy": 0.22048380225896835,
      "num_tokens": 110349101.0,
      "step": 53165
    },
    {
      "entropy": 5.44726152420044,
      "epoch": 4.82311320754717,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002908157073691233,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22362322509288787,
      "num_tokens": 110359377.0,
      "step": 53170
    },
    {
      "entropy": 5.528679895401001,
      "epoch": 4.823566763425254,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029078343082092594,
      "loss": 4.8918,
      "mean_token_accuracy": 0.22233192026615142,
      "num_tokens": 110370132.0,
      "step": 53175
    },
    {
      "entropy": 5.5091938972473145,
      "epoch": 4.824020319303338,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029075115394632325,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21668845117092134,
      "num_tokens": 110381046.0,
      "step": 53180
    },
    {
      "entropy": 5.596308040618896,
      "epoch": 4.824473875181423,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029071887674598265,
      "loss": 4.9858,
      "mean_token_accuracy": 0.2128756508231163,
      "num_tokens": 110392309.0,
      "step": 53185
    },
    {
      "entropy": 5.5867805004119875,
      "epoch": 4.824927431059507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002906865992205717,
      "loss": 5.0356,
      "mean_token_accuracy": 0.21558349579572678,
      "num_tokens": 110402464.0,
      "step": 53190
    },
    {
      "entropy": 5.40322265625,
      "epoch": 4.82538098693759,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002906543213707579,
      "loss": 4.784,
      "mean_token_accuracy": 0.23489920198917388,
      "num_tokens": 110412272.0,
      "step": 53195
    },
    {
      "entropy": 5.484185886383057,
      "epoch": 4.825834542815675,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029062204319720865,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21791450977325438,
      "num_tokens": 110423425.0,
      "step": 53200
    },
    {
      "entropy": 5.482295656204224,
      "epoch": 4.826288098693759,
      "grad_norm": 1.0,
      "learning_rate": 0.0002905897647005917,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2114737018942833,
      "num_tokens": 110434469.0,
      "step": 53205
    },
    {
      "entropy": 5.47873854637146,
      "epoch": 4.826741654571843,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002905574858815744,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2248915120959282,
      "num_tokens": 110443777.0,
      "step": 53210
    },
    {
      "entropy": 5.4891938209533695,
      "epoch": 4.827195210449927,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002905252067408242,
      "loss": 4.946,
      "mean_token_accuracy": 0.20941884815692902,
      "num_tokens": 110454481.0,
      "step": 53215
    },
    {
      "entropy": 5.5554204940795895,
      "epoch": 4.827648766328012,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002904929272790089,
      "loss": 5.0068,
      "mean_token_accuracy": 0.21611889004707335,
      "num_tokens": 110464789.0,
      "step": 53220
    },
    {
      "entropy": 5.420013523101806,
      "epoch": 4.828102322206096,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002904606474967959,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23316797316074372,
      "num_tokens": 110474276.0,
      "step": 53225
    },
    {
      "entropy": 5.522684955596924,
      "epoch": 4.82855587808418,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029042836739485273,
      "loss": 5.0238,
      "mean_token_accuracy": 0.21857748180627823,
      "num_tokens": 110486193.0,
      "step": 53230
    },
    {
      "entropy": 5.575306558609009,
      "epoch": 4.8290094339622645,
      "grad_norm": 1.234375,
      "learning_rate": 0.000290396086973847,
      "loss": 4.931,
      "mean_token_accuracy": 0.22241006195545196,
      "num_tokens": 110496619.0,
      "step": 53235
    },
    {
      "entropy": 5.5288652896881105,
      "epoch": 4.829462989840349,
      "grad_norm": 1.0,
      "learning_rate": 0.00029036380623444636,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2149721845984459,
      "num_tokens": 110507103.0,
      "step": 53240
    },
    {
      "entropy": 5.506269884109497,
      "epoch": 4.829916545718433,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002903315251773182,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22253510504961013,
      "num_tokens": 110516855.0,
      "step": 53245
    },
    {
      "entropy": 5.469615459442139,
      "epoch": 4.830370101596516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029029924380313024,
      "loss": 4.9136,
      "mean_token_accuracy": 0.2262013480067253,
      "num_tokens": 110526220.0,
      "step": 53250
    },
    {
      "entropy": 5.5293056011199955,
      "epoch": 4.830823657474601,
      "grad_norm": 1.03125,
      "learning_rate": 0.00029026696211255,
      "loss": 4.9902,
      "mean_token_accuracy": 0.2158386617898941,
      "num_tokens": 110537029.0,
      "step": 53255
    },
    {
      "entropy": 5.546607208251953,
      "epoch": 4.831277213352685,
      "grad_norm": 1.125,
      "learning_rate": 0.00029023468010624503,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22296849489212037,
      "num_tokens": 110548334.0,
      "step": 53260
    },
    {
      "entropy": 5.5496053218841555,
      "epoch": 4.831730769230769,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029020239778488303,
      "loss": 4.9517,
      "mean_token_accuracy": 0.22591616958379745,
      "num_tokens": 110557999.0,
      "step": 53265
    },
    {
      "entropy": 5.527669858932495,
      "epoch": 4.832184325108853,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002901701151491316,
      "loss": 5.0428,
      "mean_token_accuracy": 0.2164563700556755,
      "num_tokens": 110568743.0,
      "step": 53270
    },
    {
      "entropy": 5.534884595870972,
      "epoch": 4.832637880986938,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029013783219965835,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22491133064031602,
      "num_tokens": 110578961.0,
      "step": 53275
    },
    {
      "entropy": 5.558208990097046,
      "epoch": 4.833091436865022,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029010554893713086,
      "loss": 5.0125,
      "mean_token_accuracy": 0.2172553598880768,
      "num_tokens": 110590496.0,
      "step": 53280
    },
    {
      "entropy": 5.5337587833404545,
      "epoch": 4.833544992743106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002900732653622167,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22032959759235382,
      "num_tokens": 110601018.0,
      "step": 53285
    },
    {
      "entropy": 5.496733522415161,
      "epoch": 4.8339985486211905,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002900409814755836,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21779213845729828,
      "num_tokens": 110611369.0,
      "step": 53290
    },
    {
      "entropy": 5.533697509765625,
      "epoch": 4.834452104499274,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029000869727789923,
      "loss": 4.9427,
      "mean_token_accuracy": 0.22666946053504944,
      "num_tokens": 110621223.0,
      "step": 53295
    },
    {
      "entropy": 5.477897500991821,
      "epoch": 4.834905660377358,
      "grad_norm": 1.125,
      "learning_rate": 0.00028997641276983107,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2239145204424858,
      "num_tokens": 110631033.0,
      "step": 53300
    },
    {
      "entropy": 5.489819765090942,
      "epoch": 4.835359216255442,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028994412795204695,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22284831553697587,
      "num_tokens": 110642368.0,
      "step": 53305
    },
    {
      "entropy": 5.446980094909668,
      "epoch": 4.835812772133527,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002899118428252144,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21538691520690917,
      "num_tokens": 110653317.0,
      "step": 53310
    },
    {
      "entropy": 5.469147968292236,
      "epoch": 4.836266328011611,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028987955739000123,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2222746029496193,
      "num_tokens": 110663143.0,
      "step": 53315
    },
    {
      "entropy": 5.526599311828614,
      "epoch": 4.836719883889695,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000289847271647075,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22924054265022278,
      "num_tokens": 110673383.0,
      "step": 53320
    },
    {
      "entropy": 5.462620401382447,
      "epoch": 4.8371734397677795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002898149855971034,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22142033129930497,
      "num_tokens": 110683576.0,
      "step": 53325
    },
    {
      "entropy": 5.403623723983765,
      "epoch": 4.837626995645864,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00028978269924075404,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22825521230697632,
      "num_tokens": 110693986.0,
      "step": 53330
    },
    {
      "entropy": 5.411454916000366,
      "epoch": 4.838080551523948,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002897504125786948,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22597791254520416,
      "num_tokens": 110703724.0,
      "step": 53335
    },
    {
      "entropy": 5.437646532058716,
      "epoch": 4.838534107402032,
      "grad_norm": 1.125,
      "learning_rate": 0.00028971812561159324,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22373922914266586,
      "num_tokens": 110714251.0,
      "step": 53340
    },
    {
      "entropy": 5.53241548538208,
      "epoch": 4.8389876632801165,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028968583834011714,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21321135610342026,
      "num_tokens": 110724011.0,
      "step": 53345
    },
    {
      "entropy": 5.563840866088867,
      "epoch": 4.8394412191582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028965355076493405,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21935887038707733,
      "num_tokens": 110733627.0,
      "step": 53350
    },
    {
      "entropy": 5.540822982788086,
      "epoch": 4.839894775036284,
      "grad_norm": 1.0,
      "learning_rate": 0.0002896212628867118,
      "loss": 4.934,
      "mean_token_accuracy": 0.21615967601537706,
      "num_tokens": 110744514.0,
      "step": 53355
    },
    {
      "entropy": 5.52098503112793,
      "epoch": 4.8403483309143684,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00028958897470611825,
      "loss": 4.9887,
      "mean_token_accuracy": 0.2194120317697525,
      "num_tokens": 110754257.0,
      "step": 53360
    },
    {
      "entropy": 5.4577038288116455,
      "epoch": 4.840801886792453,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002895566862238208,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22374520897865297,
      "num_tokens": 110764383.0,
      "step": 53365
    },
    {
      "entropy": 5.560641670227051,
      "epoch": 4.841255442670537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002895243974404875,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21919592916965486,
      "num_tokens": 110775006.0,
      "step": 53370
    },
    {
      "entropy": 5.530359363555908,
      "epoch": 4.841708998548621,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002894921083567859,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2192254200577736,
      "num_tokens": 110784764.0,
      "step": 53375
    },
    {
      "entropy": 5.484390163421631,
      "epoch": 4.8421625544267055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002894598189733838,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22730215787887573,
      "num_tokens": 110794942.0,
      "step": 53380
    },
    {
      "entropy": 5.4386461734771725,
      "epoch": 4.84261611030479,
      "grad_norm": 1.015625,
      "learning_rate": 0.000289427529290949,
      "loss": 4.8587,
      "mean_token_accuracy": 0.2360024258494377,
      "num_tokens": 110804696.0,
      "step": 53385
    },
    {
      "entropy": 5.408174133300781,
      "epoch": 4.843069666182874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028939523931014924,
      "loss": 4.8122,
      "mean_token_accuracy": 0.23235164284706117,
      "num_tokens": 110814867.0,
      "step": 53390
    },
    {
      "entropy": 5.467607212066651,
      "epoch": 4.843523222060957,
      "grad_norm": 0.984375,
      "learning_rate": 0.00028936294903165224,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2190859705209732,
      "num_tokens": 110824738.0,
      "step": 53395
    },
    {
      "entropy": 5.480637693405152,
      "epoch": 4.843976777939043,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002893306584561257,
      "loss": 4.9768,
      "mean_token_accuracy": 0.22074196189641954,
      "num_tokens": 110834646.0,
      "step": 53400
    },
    {
      "entropy": 5.55999870300293,
      "epoch": 4.844430333817126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028929836758423763,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2170028120279312,
      "num_tokens": 110843907.0,
      "step": 53405
    },
    {
      "entropy": 5.53070650100708,
      "epoch": 4.84488388969521,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002892660764166556,
      "loss": 4.951,
      "mean_token_accuracy": 0.2208910644054413,
      "num_tokens": 110854930.0,
      "step": 53410
    },
    {
      "entropy": 5.418257141113282,
      "epoch": 4.8453374455732945,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028923378495404747,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22411495000123977,
      "num_tokens": 110865256.0,
      "step": 53415
    },
    {
      "entropy": 5.549115753173828,
      "epoch": 4.845791001451379,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028920149319708105,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21959577798843383,
      "num_tokens": 110875600.0,
      "step": 53420
    },
    {
      "entropy": 5.617954349517822,
      "epoch": 4.846244557329463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028916920114642415,
      "loss": 5.0568,
      "mean_token_accuracy": 0.2182968944311142,
      "num_tokens": 110886856.0,
      "step": 53425
    },
    {
      "entropy": 5.555357933044434,
      "epoch": 4.846698113207547,
      "grad_norm": 0.984375,
      "learning_rate": 0.00028913690880274454,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2254142567515373,
      "num_tokens": 110896908.0,
      "step": 53430
    },
    {
      "entropy": 5.484400892257691,
      "epoch": 4.8471516690856316,
      "grad_norm": 0.96875,
      "learning_rate": 0.00028910461616671004,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2257731407880783,
      "num_tokens": 110908987.0,
      "step": 53435
    },
    {
      "entropy": 5.481909513473511,
      "epoch": 4.847605224963716,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028907232323898855,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22486075311899184,
      "num_tokens": 110920036.0,
      "step": 53440
    },
    {
      "entropy": 5.538678741455078,
      "epoch": 4.8480587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028904003002024775,
      "loss": 4.9915,
      "mean_token_accuracy": 0.21645430326461793,
      "num_tokens": 110931170.0,
      "step": 53445
    },
    {
      "entropy": 5.569111108779907,
      "epoch": 4.8485123367198835,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002890077365111556,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21336396783590317,
      "num_tokens": 110942698.0,
      "step": 53450
    },
    {
      "entropy": 5.5071892738342285,
      "epoch": 4.848965892597968,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028897544271237985,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2208878830075264,
      "num_tokens": 110953400.0,
      "step": 53455
    },
    {
      "entropy": 5.543670654296875,
      "epoch": 4.849419448476052,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028894314862458844,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2233164057135582,
      "num_tokens": 110963792.0,
      "step": 53460
    },
    {
      "entropy": 5.519230365753174,
      "epoch": 4.849873004354136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028891085424844913,
      "loss": 4.9052,
      "mean_token_accuracy": 0.2188680350780487,
      "num_tokens": 110973636.0,
      "step": 53465
    },
    {
      "entropy": 5.467808818817138,
      "epoch": 4.8503265602322205,
      "grad_norm": 1.125,
      "learning_rate": 0.0002888785595846298,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22563207298517227,
      "num_tokens": 110983163.0,
      "step": 53470
    },
    {
      "entropy": 5.467434883117676,
      "epoch": 4.850780116110305,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002888462646337983,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22246607542037963,
      "num_tokens": 110993344.0,
      "step": 53475
    },
    {
      "entropy": 5.538201236724854,
      "epoch": 4.851233671988389,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028881396939662256,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2181939035654068,
      "num_tokens": 111004380.0,
      "step": 53480
    },
    {
      "entropy": 5.499354600906372,
      "epoch": 4.851687227866473,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002887816738737704,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2234104797244072,
      "num_tokens": 111015661.0,
      "step": 53485
    },
    {
      "entropy": 5.468550539016723,
      "epoch": 4.852140783744558,
      "grad_norm": 0.953125,
      "learning_rate": 0.0002887493780659097,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22484237551689149,
      "num_tokens": 111026496.0,
      "step": 53490
    },
    {
      "entropy": 5.4478003025054935,
      "epoch": 4.852594339622642,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028871708197370844,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22631847411394118,
      "num_tokens": 111036356.0,
      "step": 53495
    },
    {
      "entropy": 5.504029655456543,
      "epoch": 4.853047895500726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028868478559783433,
      "loss": 5.0001,
      "mean_token_accuracy": 0.22149582505226134,
      "num_tokens": 111045967.0,
      "step": 53500
    },
    {
      "entropy": 5.508574342727661,
      "epoch": 4.8535014513788095,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028865248893895543,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2221693605184555,
      "num_tokens": 111056459.0,
      "step": 53505
    },
    {
      "entropy": 5.48805832862854,
      "epoch": 4.853955007256894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028862019199773955,
      "loss": 4.8574,
      "mean_token_accuracy": 0.22138782888650893,
      "num_tokens": 111065901.0,
      "step": 53510
    },
    {
      "entropy": 5.468429851531982,
      "epoch": 4.854408563134978,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002885878947748546,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22285222709178926,
      "num_tokens": 111075721.0,
      "step": 53515
    },
    {
      "entropy": 5.437387466430664,
      "epoch": 4.854862119013062,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00028855559727096854,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2271849676966667,
      "num_tokens": 111086453.0,
      "step": 53520
    },
    {
      "entropy": 5.5263348579406735,
      "epoch": 4.855315674891147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028852329948674926,
      "loss": 5.0003,
      "mean_token_accuracy": 0.21554322242736818,
      "num_tokens": 111099130.0,
      "step": 53525
    },
    {
      "entropy": 5.502951431274414,
      "epoch": 4.855769230769231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002884910014228648,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22265880703926086,
      "num_tokens": 111108304.0,
      "step": 53530
    },
    {
      "entropy": 5.560021162033081,
      "epoch": 4.856222786647315,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002884587030799829,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2195548713207245,
      "num_tokens": 111118369.0,
      "step": 53535
    },
    {
      "entropy": 5.451626777648926,
      "epoch": 4.856676342525399,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028842640445877153,
      "loss": 4.897,
      "mean_token_accuracy": 0.2236127331852913,
      "num_tokens": 111128550.0,
      "step": 53540
    },
    {
      "entropy": 5.548180389404297,
      "epoch": 4.857129898403484,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028839410555989885,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20960291028022765,
      "num_tokens": 111140824.0,
      "step": 53545
    },
    {
      "entropy": 5.5037847518920895,
      "epoch": 4.857583454281567,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028836180638403257,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2201295331120491,
      "num_tokens": 111152241.0,
      "step": 53550
    },
    {
      "entropy": 5.566170740127563,
      "epoch": 4.858037010159651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002883295069318407,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21468812972307205,
      "num_tokens": 111161897.0,
      "step": 53555
    },
    {
      "entropy": 5.518864345550537,
      "epoch": 4.8584905660377355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002882972072039914,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22234005331993104,
      "num_tokens": 111172586.0,
      "step": 53560
    },
    {
      "entropy": 5.530705785751342,
      "epoch": 4.85894412191582,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028826490720115224,
      "loss": 5.0375,
      "mean_token_accuracy": 0.21549375206232071,
      "num_tokens": 111182734.0,
      "step": 53565
    },
    {
      "entropy": 5.447303199768067,
      "epoch": 4.859397677793904,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028823260692399156,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23129567950963975,
      "num_tokens": 111193551.0,
      "step": 53570
    },
    {
      "entropy": 5.464043712615966,
      "epoch": 4.859851233671988,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002882003063731771,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2232265591621399,
      "num_tokens": 111202604.0,
      "step": 53575
    },
    {
      "entropy": 5.4702291011810305,
      "epoch": 4.860304789550073,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00028816800554937703,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22917802333831788,
      "num_tokens": 111213675.0,
      "step": 53580
    },
    {
      "entropy": 5.5303037643432615,
      "epoch": 4.860758345428157,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028813570445325915,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2294157013297081,
      "num_tokens": 111223354.0,
      "step": 53585
    },
    {
      "entropy": 5.505362319946289,
      "epoch": 4.861211901306241,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028810340308549164,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22760512679815292,
      "num_tokens": 111232865.0,
      "step": 53590
    },
    {
      "entropy": 5.4311363220214846,
      "epoch": 4.861665457184325,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028807110144674243,
      "loss": 4.8771,
      "mean_token_accuracy": 0.23055828660726546,
      "num_tokens": 111242924.0,
      "step": 53595
    },
    {
      "entropy": 5.45285325050354,
      "epoch": 4.86211901306241,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002880387995376794,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23394269198179246,
      "num_tokens": 111252921.0,
      "step": 53600
    },
    {
      "entropy": 5.520256805419922,
      "epoch": 4.862572568940493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002880064973589708,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21900326609611512,
      "num_tokens": 111263176.0,
      "step": 53605
    },
    {
      "entropy": 5.495283317565918,
      "epoch": 4.863026124818577,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028797419491128446,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21684139966964722,
      "num_tokens": 111273043.0,
      "step": 53610
    },
    {
      "entropy": 5.49081563949585,
      "epoch": 4.863479680696662,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028794189219528846,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22229012101888657,
      "num_tokens": 111282720.0,
      "step": 53615
    },
    {
      "entropy": 5.425160312652588,
      "epoch": 4.863933236574746,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028790958921165086,
      "loss": 4.9187,
      "mean_token_accuracy": 0.20915600210428237,
      "num_tokens": 111293946.0,
      "step": 53620
    },
    {
      "entropy": 5.532509088516235,
      "epoch": 4.86438679245283,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002878772859610396,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21759513318538665,
      "num_tokens": 111305052.0,
      "step": 53625
    },
    {
      "entropy": 5.5392045974731445,
      "epoch": 4.864840348330914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028784498244412284,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2220151498913765,
      "num_tokens": 111314216.0,
      "step": 53630
    },
    {
      "entropy": 5.442796802520752,
      "epoch": 4.865293904208999,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028781267866156854,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22607042044401168,
      "num_tokens": 111324939.0,
      "step": 53635
    },
    {
      "entropy": 5.517685270309448,
      "epoch": 4.865747460087083,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00028778037461404477,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2211136132478714,
      "num_tokens": 111334908.0,
      "step": 53640
    },
    {
      "entropy": 5.48081636428833,
      "epoch": 4.866201015965167,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028774807030221966,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2260892406105995,
      "num_tokens": 111345560.0,
      "step": 53645
    },
    {
      "entropy": 5.488665771484375,
      "epoch": 4.8666545718432515,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002877157657267611,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2284022703766823,
      "num_tokens": 111355955.0,
      "step": 53650
    },
    {
      "entropy": 5.526673412322998,
      "epoch": 4.867108127721336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028768346088833737,
      "loss": 4.9894,
      "mean_token_accuracy": 0.22553099542856217,
      "num_tokens": 111365567.0,
      "step": 53655
    },
    {
      "entropy": 5.413854551315308,
      "epoch": 4.867561683599419,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002876511557876164,
      "loss": 4.8439,
      "mean_token_accuracy": 0.2270178109407425,
      "num_tokens": 111375740.0,
      "step": 53660
    },
    {
      "entropy": 5.490636873245239,
      "epoch": 4.868015239477503,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028761885042526635,
      "loss": 4.896,
      "mean_token_accuracy": 0.22709425687789916,
      "num_tokens": 111385593.0,
      "step": 53665
    },
    {
      "entropy": 5.545875549316406,
      "epoch": 4.868468795355588,
      "grad_norm": 1.125,
      "learning_rate": 0.00028758654480195517,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2127946600317955,
      "num_tokens": 111395731.0,
      "step": 53670
    },
    {
      "entropy": 5.535590553283692,
      "epoch": 4.868922351233672,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002875542389183511,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2260356217622757,
      "num_tokens": 111406140.0,
      "step": 53675
    },
    {
      "entropy": 5.546180486679077,
      "epoch": 4.869375907111756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028752193277512217,
      "loss": 4.958,
      "mean_token_accuracy": 0.22034520506858826,
      "num_tokens": 111416853.0,
      "step": 53680
    },
    {
      "entropy": 5.516473245620728,
      "epoch": 4.86982946298984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002874896263729364,
      "loss": 4.8736,
      "mean_token_accuracy": 0.22610962390899658,
      "num_tokens": 111427043.0,
      "step": 53685
    },
    {
      "entropy": 5.560755586624145,
      "epoch": 4.870283018867925,
      "grad_norm": 1.125,
      "learning_rate": 0.0002874573197124621,
      "loss": 4.9838,
      "mean_token_accuracy": 0.22054844796657563,
      "num_tokens": 111436791.0,
      "step": 53690
    },
    {
      "entropy": 5.424797821044922,
      "epoch": 4.870736574746009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002874250127943672,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22976084202528,
      "num_tokens": 111446603.0,
      "step": 53695
    },
    {
      "entropy": 5.415288877487183,
      "epoch": 4.871190130624093,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028739270561931984,
      "loss": 4.8258,
      "mean_token_accuracy": 0.2327849179506302,
      "num_tokens": 111456942.0,
      "step": 53700
    },
    {
      "entropy": 5.502159929275512,
      "epoch": 4.871643686502177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028736039818798813,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21901227235794068,
      "num_tokens": 111467671.0,
      "step": 53705
    },
    {
      "entropy": 5.555570077896118,
      "epoch": 4.872097242380261,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002873280905010403,
      "loss": 4.9829,
      "mean_token_accuracy": 0.2145502209663391,
      "num_tokens": 111476495.0,
      "step": 53710
    },
    {
      "entropy": 5.455052757263184,
      "epoch": 4.872550798258345,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028729578255914446,
      "loss": 4.7642,
      "mean_token_accuracy": 0.23636919558048247,
      "num_tokens": 111487491.0,
      "step": 53715
    },
    {
      "entropy": 5.462331533432007,
      "epoch": 4.873004354136429,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002872634743629686,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2219252571463585,
      "num_tokens": 111497970.0,
      "step": 53720
    },
    {
      "entropy": 5.4763871192932125,
      "epoch": 4.873457910014514,
      "grad_norm": 0.984375,
      "learning_rate": 0.000287231165913181,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21994804590940475,
      "num_tokens": 111507635.0,
      "step": 53725
    },
    {
      "entropy": 5.453107213973999,
      "epoch": 4.873911465892598,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002871988572104498,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22866019755601882,
      "num_tokens": 111516751.0,
      "step": 53730
    },
    {
      "entropy": 5.492450332641601,
      "epoch": 4.874365021770682,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028716654825544316,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2220484048128128,
      "num_tokens": 111527748.0,
      "step": 53735
    },
    {
      "entropy": 5.548152828216553,
      "epoch": 4.8748185776487665,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002871342390488291,
      "loss": 4.923,
      "mean_token_accuracy": 0.21587084382772445,
      "num_tokens": 111537718.0,
      "step": 53740
    },
    {
      "entropy": 5.403615140914917,
      "epoch": 4.875272133526851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028710192959127597,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23522300124168397,
      "num_tokens": 111546631.0,
      "step": 53745
    },
    {
      "entropy": 5.46729850769043,
      "epoch": 4.875725689404935,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002870696198834518,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22216736972332002,
      "num_tokens": 111557276.0,
      "step": 53750
    },
    {
      "entropy": 5.494259786605835,
      "epoch": 4.876179245283019,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028703730992602486,
      "loss": 4.8891,
      "mean_token_accuracy": 0.22493528872728347,
      "num_tokens": 111567146.0,
      "step": 53755
    },
    {
      "entropy": 5.534890460968017,
      "epoch": 4.876632801161103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028700499971966327,
      "loss": 4.9877,
      "mean_token_accuracy": 0.21771259605884552,
      "num_tokens": 111576887.0,
      "step": 53760
    },
    {
      "entropy": 5.567113399505615,
      "epoch": 4.877086357039187,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028697268926503527,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21017972230911255,
      "num_tokens": 111588212.0,
      "step": 53765
    },
    {
      "entropy": 5.483321046829223,
      "epoch": 4.877539912917271,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028694037856280894,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2236213505268097,
      "num_tokens": 111598308.0,
      "step": 53770
    },
    {
      "entropy": 5.506678676605224,
      "epoch": 4.8779934687953554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002869080676136525,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2226239949464798,
      "num_tokens": 111608252.0,
      "step": 53775
    },
    {
      "entropy": 5.50646743774414,
      "epoch": 4.87844702467344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002868757564182343,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21775002777576447,
      "num_tokens": 111619555.0,
      "step": 53780
    },
    {
      "entropy": 5.43842716217041,
      "epoch": 4.878900580551524,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002868434449772223,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22868140488862992,
      "num_tokens": 111629578.0,
      "step": 53785
    },
    {
      "entropy": 5.520170783996582,
      "epoch": 4.879354136429608,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002868111332912849,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21629922091960907,
      "num_tokens": 111639389.0,
      "step": 53790
    },
    {
      "entropy": 5.602502012252808,
      "epoch": 4.8798076923076925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002867788213610903,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21706010699272155,
      "num_tokens": 111649855.0,
      "step": 53795
    },
    {
      "entropy": 5.5797138690948485,
      "epoch": 4.880261248185777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002867465091873066,
      "loss": 5.0112,
      "mean_token_accuracy": 0.2195116639137268,
      "num_tokens": 111660315.0,
      "step": 53800
    },
    {
      "entropy": 5.527353715896607,
      "epoch": 4.880714804063861,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002867141967706021,
      "loss": 4.964,
      "mean_token_accuracy": 0.21600376218557357,
      "num_tokens": 111669883.0,
      "step": 53805
    },
    {
      "entropy": 5.457800722122192,
      "epoch": 4.881168359941945,
      "grad_norm": 1.0,
      "learning_rate": 0.00028668188411164506,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21693479120731354,
      "num_tokens": 111680598.0,
      "step": 53810
    },
    {
      "entropy": 5.536665391921997,
      "epoch": 4.881621915820029,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028664957121110374,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2223975971341133,
      "num_tokens": 111689805.0,
      "step": 53815
    },
    {
      "entropy": 5.526259136199951,
      "epoch": 4.882075471698113,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00028661725806964624,
      "loss": 4.8545,
      "mean_token_accuracy": 0.2298549383878708,
      "num_tokens": 111700313.0,
      "step": 53820
    },
    {
      "entropy": 5.551016569137573,
      "epoch": 4.882529027576197,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002865849446879409,
      "loss": 4.9735,
      "mean_token_accuracy": 0.22339250892400742,
      "num_tokens": 111711301.0,
      "step": 53825
    },
    {
      "entropy": 5.4724711894989015,
      "epoch": 4.8829825834542815,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002865526310666559,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22118804901838302,
      "num_tokens": 111720800.0,
      "step": 53830
    },
    {
      "entropy": 5.480409336090088,
      "epoch": 4.883436139332366,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00028652031720645963,
      "loss": 4.837,
      "mean_token_accuracy": 0.22383893728256227,
      "num_tokens": 111731665.0,
      "step": 53835
    },
    {
      "entropy": 5.492449188232422,
      "epoch": 4.88388969521045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002864880031080202,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22362476140260695,
      "num_tokens": 111742149.0,
      "step": 53840
    },
    {
      "entropy": 5.511508226394653,
      "epoch": 4.884343251088534,
      "grad_norm": 0.953125,
      "learning_rate": 0.00028645568877200603,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22956591546535493,
      "num_tokens": 111753089.0,
      "step": 53845
    },
    {
      "entropy": 5.461369466781616,
      "epoch": 4.8847968069666186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002864233741990852,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2284594178199768,
      "num_tokens": 111762937.0,
      "step": 53850
    },
    {
      "entropy": 5.43382887840271,
      "epoch": 4.885250362844703,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002863910593899261,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2350654199719429,
      "num_tokens": 111772920.0,
      "step": 53855
    },
    {
      "entropy": 5.469610929489136,
      "epoch": 4.885703918722786,
      "grad_norm": 1.015625,
      "learning_rate": 0.000286358744345197,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22244598865509033,
      "num_tokens": 111783117.0,
      "step": 53860
    },
    {
      "entropy": 5.485313987731933,
      "epoch": 4.8861574746008705,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0002863264290655662,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21588664799928664,
      "num_tokens": 111795033.0,
      "step": 53865
    },
    {
      "entropy": 5.4740385055542,
      "epoch": 4.886611030478955,
      "grad_norm": 1.0,
      "learning_rate": 0.00028629411355170193,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21807047873735427,
      "num_tokens": 111805156.0,
      "step": 53870
    },
    {
      "entropy": 5.556977844238281,
      "epoch": 4.887064586357039,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028626179780427246,
      "loss": 4.9849,
      "mean_token_accuracy": 0.21690288335084915,
      "num_tokens": 111815363.0,
      "step": 53875
    },
    {
      "entropy": 5.549709272384644,
      "epoch": 4.887518142235123,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028622948182394623,
      "loss": 4.9468,
      "mean_token_accuracy": 0.22135339826345443,
      "num_tokens": 111824983.0,
      "step": 53880
    },
    {
      "entropy": 5.449427795410156,
      "epoch": 4.8879716981132075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028619716561139143,
      "loss": 4.932,
      "mean_token_accuracy": 0.21982220858335494,
      "num_tokens": 111836468.0,
      "step": 53885
    },
    {
      "entropy": 5.503212022781372,
      "epoch": 4.888425253991292,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002861648491672764,
      "loss": 4.9529,
      "mean_token_accuracy": 0.22237562388181686,
      "num_tokens": 111847659.0,
      "step": 53890
    },
    {
      "entropy": 5.572929954528808,
      "epoch": 4.888878809869376,
      "grad_norm": 1.0,
      "learning_rate": 0.00028613253249226944,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2150169938802719,
      "num_tokens": 111858742.0,
      "step": 53895
    },
    {
      "entropy": 5.540292835235595,
      "epoch": 4.88933236574746,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00028610021558703885,
      "loss": 4.9547,
      "mean_token_accuracy": 0.22229746729135513,
      "num_tokens": 111869692.0,
      "step": 53900
    },
    {
      "entropy": 5.482226324081421,
      "epoch": 4.889785921625545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028606789845225295,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22444059699773788,
      "num_tokens": 111880532.0,
      "step": 53905
    },
    {
      "entropy": 5.542838001251221,
      "epoch": 4.890239477503629,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028603558108858016,
      "loss": 5.0129,
      "mean_token_accuracy": 0.21617412865161895,
      "num_tokens": 111891178.0,
      "step": 53910
    },
    {
      "entropy": 5.557167100906372,
      "epoch": 4.890693033381712,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002860032634966887,
      "loss": 4.9643,
      "mean_token_accuracy": 0.22316309213638305,
      "num_tokens": 111902043.0,
      "step": 53915
    },
    {
      "entropy": 5.559648084640503,
      "epoch": 4.8911465892597965,
      "grad_norm": 1.0625,
      "learning_rate": 0.000285970945677247,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22651675790548326,
      "num_tokens": 111912170.0,
      "step": 53920
    },
    {
      "entropy": 5.4224954605102536,
      "epoch": 4.891600145137881,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00028593862763092327,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22774345576763153,
      "num_tokens": 111922415.0,
      "step": 53925
    },
    {
      "entropy": 5.582417392730713,
      "epoch": 4.892053701015965,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028590630935838596,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21370297223329543,
      "num_tokens": 111933338.0,
      "step": 53930
    },
    {
      "entropy": 5.503468132019043,
      "epoch": 4.892507256894049,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002858739908603034,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22623978108167647,
      "num_tokens": 111942225.0,
      "step": 53935
    },
    {
      "entropy": 5.488671064376831,
      "epoch": 4.892960812772134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002858416721373439,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2278818368911743,
      "num_tokens": 111953356.0,
      "step": 53940
    },
    {
      "entropy": 5.456784343719482,
      "epoch": 4.893414368650218,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028580935319017593,
      "loss": 4.8289,
      "mean_token_accuracy": 0.22111749202013015,
      "num_tokens": 111964275.0,
      "step": 53945
    },
    {
      "entropy": 5.479333353042603,
      "epoch": 4.893867924528302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028577703401946777,
      "loss": 4.894,
      "mean_token_accuracy": 0.22689733505249024,
      "num_tokens": 111974006.0,
      "step": 53950
    },
    {
      "entropy": 5.553197193145752,
      "epoch": 4.894321480406386,
      "grad_norm": 1.0,
      "learning_rate": 0.00028574471462588776,
      "loss": 5.0404,
      "mean_token_accuracy": 0.2151483729481697,
      "num_tokens": 111985085.0,
      "step": 53955
    },
    {
      "entropy": 5.585968017578125,
      "epoch": 4.89477503628447,
      "grad_norm": 1.125,
      "learning_rate": 0.0002857123950101043,
      "loss": 5.0145,
      "mean_token_accuracy": 0.21336821615695953,
      "num_tokens": 111995162.0,
      "step": 53960
    },
    {
      "entropy": 5.5790266513824465,
      "epoch": 4.895228592162555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002856800751727858,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21886659413576126,
      "num_tokens": 112005625.0,
      "step": 53965
    },
    {
      "entropy": 5.466133069992066,
      "epoch": 4.895682148040638,
      "grad_norm": 1.0,
      "learning_rate": 0.00028564775511460066,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22935950458049775,
      "num_tokens": 112015977.0,
      "step": 53970
    },
    {
      "entropy": 5.493181276321411,
      "epoch": 4.8961357039187225,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002856154348362173,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22061038315296172,
      "num_tokens": 112025655.0,
      "step": 53975
    },
    {
      "entropy": 5.459987163543701,
      "epoch": 4.896589259796807,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028558311433830386,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22969314455986023,
      "num_tokens": 112036986.0,
      "step": 53980
    },
    {
      "entropy": 5.490080261230469,
      "epoch": 4.897042815674891,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000285550793621529,
      "loss": 4.9256,
      "mean_token_accuracy": 0.22587249279022217,
      "num_tokens": 112047541.0,
      "step": 53985
    },
    {
      "entropy": 5.50920352935791,
      "epoch": 4.897496371552975,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002855184726865611,
      "loss": 4.886,
      "mean_token_accuracy": 0.2199341833591461,
      "num_tokens": 112058125.0,
      "step": 53990
    },
    {
      "entropy": 5.545265245437622,
      "epoch": 4.89794992743106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028548615153406844,
      "loss": 4.9557,
      "mean_token_accuracy": 0.21945470571517944,
      "num_tokens": 112068712.0,
      "step": 53995
    },
    {
      "entropy": 5.456097555160523,
      "epoch": 4.898403483309144,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00028545383016471954,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2194112718105316,
      "num_tokens": 112078529.0,
      "step": 54000
    },
    {
      "epoch": 4.898403483309144,
      "eval_entropy": 5.271888157361233,
      "eval_loss": 4.999665260314941,
      "eval_mean_token_accuracy": 0.22641565401393635,
      "eval_num_tokens": 112078529.0,
      "eval_runtime": 26.7217,
      "eval_samples_per_second": 1323.267,
      "eval_steps_per_second": 165.408,
      "step": 54000
    },
    {
      "entropy": 5.575981950759887,
      "epoch": 4.898857039187228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002854215085791826,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21349867582321166,
      "num_tokens": 112088814.0,
      "step": 54005
    },
    {
      "entropy": 5.503803396224976,
      "epoch": 4.899310595065312,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028538918677812643,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2289568930864334,
      "num_tokens": 112099406.0,
      "step": 54010
    },
    {
      "entropy": 5.507465553283692,
      "epoch": 4.899764150943396,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00028535686476221915,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21903334707021713,
      "num_tokens": 112110762.0,
      "step": 54015
    },
    {
      "entropy": 5.613582611083984,
      "epoch": 4.90021770682148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002853245425321292,
      "loss": 5.0922,
      "mean_token_accuracy": 0.20801510363817216,
      "num_tokens": 112120636.0,
      "step": 54020
    },
    {
      "entropy": 5.492752742767334,
      "epoch": 4.900671262699564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028529222008852513,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22706396728754044,
      "num_tokens": 112130756.0,
      "step": 54025
    },
    {
      "entropy": 5.555766201019287,
      "epoch": 4.901124818577649,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002852598974320753,
      "loss": 4.9547,
      "mean_token_accuracy": 0.22894134223461152,
      "num_tokens": 112140881.0,
      "step": 54030
    },
    {
      "entropy": 5.492450714111328,
      "epoch": 4.901578374455733,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028522757456344824,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22385923564434052,
      "num_tokens": 112152337.0,
      "step": 54035
    },
    {
      "entropy": 5.543563604354858,
      "epoch": 4.902031930333817,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002851952514833123,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21678255647420883,
      "num_tokens": 112164199.0,
      "step": 54040
    },
    {
      "entropy": 5.492858457565307,
      "epoch": 4.902485486211901,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028516292819233597,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22488367110490798,
      "num_tokens": 112173594.0,
      "step": 54045
    },
    {
      "entropy": 5.473372125625611,
      "epoch": 4.902939042089986,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028513060469118767,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2229302003979683,
      "num_tokens": 112183261.0,
      "step": 54050
    },
    {
      "entropy": 5.45235276222229,
      "epoch": 4.90339259796807,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028509828098053594,
      "loss": 4.8713,
      "mean_token_accuracy": 0.229483525454998,
      "num_tokens": 112193506.0,
      "step": 54055
    },
    {
      "entropy": 5.462916040420533,
      "epoch": 4.903846153846154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002850659570610491,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2202925354242325,
      "num_tokens": 112204665.0,
      "step": 54060
    },
    {
      "entropy": 5.49705958366394,
      "epoch": 4.9042997097242385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002850336329333958,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22681606262922288,
      "num_tokens": 112214933.0,
      "step": 54065
    },
    {
      "entropy": 5.5003821849823,
      "epoch": 4.904753265602322,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002850013085982443,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2275715634226799,
      "num_tokens": 112224525.0,
      "step": 54070
    },
    {
      "entropy": 5.475374555587768,
      "epoch": 4.905206821480406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028496898405626326,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22789170295000077,
      "num_tokens": 112234363.0,
      "step": 54075
    },
    {
      "entropy": 5.474161148071289,
      "epoch": 4.90566037735849,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028493665930812113,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22581566274166107,
      "num_tokens": 112245721.0,
      "step": 54080
    },
    {
      "entropy": 5.509126615524292,
      "epoch": 4.906113933236575,
      "grad_norm": 1.125,
      "learning_rate": 0.0002849043343544863,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22685883939266205,
      "num_tokens": 112255775.0,
      "step": 54085
    },
    {
      "entropy": 5.5137864112854,
      "epoch": 4.906567489114659,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002848720091960273,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21401074379682541,
      "num_tokens": 112266049.0,
      "step": 54090
    },
    {
      "entropy": 5.476232051849365,
      "epoch": 4.907021044992743,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028483968383341264,
      "loss": 4.8951,
      "mean_token_accuracy": 0.21591277718544005,
      "num_tokens": 112276663.0,
      "step": 54095
    },
    {
      "entropy": 5.472217655181884,
      "epoch": 4.907474600870827,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002848073582673109,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2266436904668808,
      "num_tokens": 112287065.0,
      "step": 54100
    },
    {
      "entropy": 5.511583948135376,
      "epoch": 4.907928156748912,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002847750324983903,
      "loss": 4.9199,
      "mean_token_accuracy": 0.21771412938833237,
      "num_tokens": 112297053.0,
      "step": 54105
    },
    {
      "entropy": 5.484909439086914,
      "epoch": 4.908381712626996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028474270652731966,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22319669723510743,
      "num_tokens": 112306001.0,
      "step": 54110
    },
    {
      "entropy": 5.522385692596435,
      "epoch": 4.908835268505079,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028471038035476735,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22891612648963927,
      "num_tokens": 112315865.0,
      "step": 54115
    },
    {
      "entropy": 5.559567880630493,
      "epoch": 4.9092888243831645,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002846780539814019,
      "loss": 5.0163,
      "mean_token_accuracy": 0.21464449912309647,
      "num_tokens": 112326768.0,
      "step": 54120
    },
    {
      "entropy": 5.446709537506104,
      "epoch": 4.909742380261248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028464572740789177,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23256117403507232,
      "num_tokens": 112337272.0,
      "step": 54125
    },
    {
      "entropy": 5.5605635166168215,
      "epoch": 4.910195936139332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002846134006349057,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21611292213201522,
      "num_tokens": 112346288.0,
      "step": 54130
    },
    {
      "entropy": 5.425166845321655,
      "epoch": 4.910649492017416,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002845810736631119,
      "loss": 4.7814,
      "mean_token_accuracy": 0.22894555181264878,
      "num_tokens": 112356829.0,
      "step": 54135
    },
    {
      "entropy": 5.422042465209961,
      "epoch": 4.911103047895501,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028454874649317915,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22848352342844008,
      "num_tokens": 112367301.0,
      "step": 54140
    },
    {
      "entropy": 5.540562295913697,
      "epoch": 4.911556603773585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028451641912577587,
      "loss": 4.9064,
      "mean_token_accuracy": 0.2253034606575966,
      "num_tokens": 112377638.0,
      "step": 54145
    },
    {
      "entropy": 5.538688135147095,
      "epoch": 4.912010159651669,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00028448409156157055,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2279538780450821,
      "num_tokens": 112387629.0,
      "step": 54150
    },
    {
      "entropy": 5.515573740005493,
      "epoch": 4.9124637155297535,
      "grad_norm": 1.125,
      "learning_rate": 0.0002844517638012318,
      "loss": 4.9659,
      "mean_token_accuracy": 0.22124159038066865,
      "num_tokens": 112397216.0,
      "step": 54155
    },
    {
      "entropy": 5.525775480270386,
      "epoch": 4.912917271407838,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028441943584542827,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22440945655107497,
      "num_tokens": 112407438.0,
      "step": 54160
    },
    {
      "entropy": 5.493718433380127,
      "epoch": 4.913370827285922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028438710769482836,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2349713310599327,
      "num_tokens": 112417328.0,
      "step": 54165
    },
    {
      "entropy": 5.509230661392212,
      "epoch": 4.913824383164005,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002843547793501007,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2228181004524231,
      "num_tokens": 112427711.0,
      "step": 54170
    },
    {
      "entropy": 5.501843595504761,
      "epoch": 4.91427793904209,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002843224508119137,
      "loss": 4.925,
      "mean_token_accuracy": 0.22424444407224656,
      "num_tokens": 112438953.0,
      "step": 54175
    },
    {
      "entropy": 5.543777322769165,
      "epoch": 4.914731494920174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002842901220809362,
      "loss": 4.9626,
      "mean_token_accuracy": 0.22525679022073747,
      "num_tokens": 112449029.0,
      "step": 54180
    },
    {
      "entropy": 5.436410045623779,
      "epoch": 4.915185050798258,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002842577931578365,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23122911006212235,
      "num_tokens": 112458790.0,
      "step": 54185
    },
    {
      "entropy": 5.438695478439331,
      "epoch": 4.9156386066763424,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002842254640432834,
      "loss": 4.9003,
      "mean_token_accuracy": 0.2200777679681778,
      "num_tokens": 112469723.0,
      "step": 54190
    },
    {
      "entropy": 5.5008885860443115,
      "epoch": 4.916092162554427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028419313473794516,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2277955949306488,
      "num_tokens": 112480448.0,
      "step": 54195
    },
    {
      "entropy": 5.489368200302124,
      "epoch": 4.916545718432511,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028416080524249073,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2271230101585388,
      "num_tokens": 112490107.0,
      "step": 54200
    },
    {
      "entropy": 5.49259386062622,
      "epoch": 4.916999274310595,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028412847555758835,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22164175659418106,
      "num_tokens": 112499670.0,
      "step": 54205
    },
    {
      "entropy": 5.457855367660523,
      "epoch": 4.9174528301886795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002840961456839069,
      "loss": 4.8478,
      "mean_token_accuracy": 0.22878511548042296,
      "num_tokens": 112509087.0,
      "step": 54210
    },
    {
      "entropy": 5.477168941497803,
      "epoch": 4.917906386066764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002840638156221148,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22706625312566758,
      "num_tokens": 112518506.0,
      "step": 54215
    },
    {
      "entropy": 5.526874542236328,
      "epoch": 4.918359941944848,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028403148537288063,
      "loss": 4.9284,
      "mean_token_accuracy": 0.22078175991773605,
      "num_tokens": 112528474.0,
      "step": 54220
    },
    {
      "entropy": 5.5161339282989506,
      "epoch": 4.918813497822931,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002839991549368731,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2175062730908394,
      "num_tokens": 112538513.0,
      "step": 54225
    },
    {
      "entropy": 5.573317384719848,
      "epoch": 4.919267053701016,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028396682431476074,
      "loss": 4.9988,
      "mean_token_accuracy": 0.2182699427008629,
      "num_tokens": 112548228.0,
      "step": 54230
    },
    {
      "entropy": 5.507677030563355,
      "epoch": 4.9197206095791,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028393449350721216,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22419071495532988,
      "num_tokens": 112558218.0,
      "step": 54235
    },
    {
      "entropy": 5.515549945831299,
      "epoch": 4.920174165457184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000283902162514896,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22003841549158096,
      "num_tokens": 112568103.0,
      "step": 54240
    },
    {
      "entropy": 5.441576671600342,
      "epoch": 4.9206277213352685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002838698313384809,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22775764614343644,
      "num_tokens": 112577729.0,
      "step": 54245
    },
    {
      "entropy": 5.530253839492798,
      "epoch": 4.921081277213353,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002838374999786353,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21973712146282195,
      "num_tokens": 112589103.0,
      "step": 54250
    },
    {
      "entropy": 5.483915615081787,
      "epoch": 4.921534833091437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028380516843602797,
      "loss": 4.8481,
      "mean_token_accuracy": 0.23137497454881667,
      "num_tokens": 112598935.0,
      "step": 54255
    },
    {
      "entropy": 5.519485569000244,
      "epoch": 4.921988388969521,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028377283671132756,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22575659304857254,
      "num_tokens": 112609837.0,
      "step": 54260
    },
    {
      "entropy": 5.517656230926514,
      "epoch": 4.9224419448476056,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002837405048052027,
      "loss": 4.9429,
      "mean_token_accuracy": 0.22429914474487306,
      "num_tokens": 112620159.0,
      "step": 54265
    },
    {
      "entropy": 5.45650200843811,
      "epoch": 4.922895500725689,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028370817271832193,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22657536417245866,
      "num_tokens": 112630779.0,
      "step": 54270
    },
    {
      "entropy": 5.427998733520508,
      "epoch": 4.923349056603773,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002836758404513539,
      "loss": 4.879,
      "mean_token_accuracy": 0.22296125888824464,
      "num_tokens": 112640702.0,
      "step": 54275
    },
    {
      "entropy": 5.540637350082397,
      "epoch": 4.9238026124818575,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002836435080049673,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22258661538362504,
      "num_tokens": 112651294.0,
      "step": 54280
    },
    {
      "entropy": 5.557561159133911,
      "epoch": 4.924256168359942,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002836111753798308,
      "loss": 4.9689,
      "mean_token_accuracy": 0.2147477760910988,
      "num_tokens": 112661570.0,
      "step": 54285
    },
    {
      "entropy": 5.49298882484436,
      "epoch": 4.924709724238026,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028357884257661285,
      "loss": 4.9471,
      "mean_token_accuracy": 0.22419185191392899,
      "num_tokens": 112671996.0,
      "step": 54290
    },
    {
      "entropy": 5.465556335449219,
      "epoch": 4.92516328011611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002835465095959824,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23151538223028184,
      "num_tokens": 112681661.0,
      "step": 54295
    },
    {
      "entropy": 5.57258973121643,
      "epoch": 4.9256168359941945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002835141764386079,
      "loss": 4.9752,
      "mean_token_accuracy": 0.22478580325841904,
      "num_tokens": 112692026.0,
      "step": 54300
    },
    {
      "entropy": 5.574635028839111,
      "epoch": 4.926070391872279,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000283481843105158,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2155161455273628,
      "num_tokens": 112703011.0,
      "step": 54305
    },
    {
      "entropy": 5.550316619873047,
      "epoch": 4.926523947750363,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002834495095963015,
      "loss": 4.9724,
      "mean_token_accuracy": 0.21721527874469757,
      "num_tokens": 112712687.0,
      "step": 54310
    },
    {
      "entropy": 5.577332305908203,
      "epoch": 4.926977503628447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000283417175912707,
      "loss": 4.9721,
      "mean_token_accuracy": 0.22375713735818864,
      "num_tokens": 112723422.0,
      "step": 54315
    },
    {
      "entropy": 5.504968309402466,
      "epoch": 4.927431059506532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002833848420550432,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22537158131599427,
      "num_tokens": 112733793.0,
      "step": 54320
    },
    {
      "entropy": 5.491259622573852,
      "epoch": 4.927884615384615,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00028335250802397863,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22936502248048782,
      "num_tokens": 112744175.0,
      "step": 54325
    },
    {
      "entropy": 5.551631212234497,
      "epoch": 4.928338171262699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028332017382018203,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21733730137348176,
      "num_tokens": 112754484.0,
      "step": 54330
    },
    {
      "entropy": 5.416099929809571,
      "epoch": 4.9287917271407835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002832878394443222,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24158648848533631,
      "num_tokens": 112764245.0,
      "step": 54335
    },
    {
      "entropy": 5.470212602615357,
      "epoch": 4.929245283018868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028325550489706773,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22538939714431763,
      "num_tokens": 112773629.0,
      "step": 54340
    },
    {
      "entropy": 5.4659630298614506,
      "epoch": 4.929698838896952,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002832231701790873,
      "loss": 4.8884,
      "mean_token_accuracy": 0.220974338054657,
      "num_tokens": 112784091.0,
      "step": 54345
    },
    {
      "entropy": 5.465558385848999,
      "epoch": 4.930152394775036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028319083529104966,
      "loss": 4.8428,
      "mean_token_accuracy": 0.2228279948234558,
      "num_tokens": 112794116.0,
      "step": 54350
    },
    {
      "entropy": 5.492745447158813,
      "epoch": 4.930605950653121,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028315850023362345,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22196406424045562,
      "num_tokens": 112804717.0,
      "step": 54355
    },
    {
      "entropy": 5.459794759750366,
      "epoch": 4.931059506531205,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00028312616500747727,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22348054200410844,
      "num_tokens": 112815384.0,
      "step": 54360
    },
    {
      "entropy": 5.509565019607544,
      "epoch": 4.931513062409289,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028309382961328003,
      "loss": 4.9553,
      "mean_token_accuracy": 0.22136979103088378,
      "num_tokens": 112825813.0,
      "step": 54365
    },
    {
      "entropy": 5.461468601226807,
      "epoch": 4.931966618287373,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028306149405170035,
      "loss": 4.86,
      "mean_token_accuracy": 0.2329668089747429,
      "num_tokens": 112836032.0,
      "step": 54370
    },
    {
      "entropy": 5.459139013290406,
      "epoch": 4.932420174165458,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002830291583234068,
      "loss": 4.9082,
      "mean_token_accuracy": 0.2231154128909111,
      "num_tokens": 112846042.0,
      "step": 54375
    },
    {
      "entropy": 5.495593786239624,
      "epoch": 4.932873730043541,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00028299682242906825,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21921909749507903,
      "num_tokens": 112856409.0,
      "step": 54380
    },
    {
      "entropy": 5.541771650314331,
      "epoch": 4.933327285921625,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00028296448636935345,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21636230647563934,
      "num_tokens": 112866788.0,
      "step": 54385
    },
    {
      "entropy": 5.433371210098267,
      "epoch": 4.9337808417997095,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028293215014493095,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22531136870384216,
      "num_tokens": 112876670.0,
      "step": 54390
    },
    {
      "entropy": 5.5159941673278805,
      "epoch": 4.934234397677794,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002828998137564697,
      "loss": 4.9337,
      "mean_token_accuracy": 0.2191131144762039,
      "num_tokens": 112887337.0,
      "step": 54395
    },
    {
      "entropy": 5.4426515102386475,
      "epoch": 4.934687953555878,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002828674772046382,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2153196394443512,
      "num_tokens": 112897693.0,
      "step": 54400
    },
    {
      "entropy": 5.488088798522949,
      "epoch": 4.935141509433962,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002828351404901052,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2133613646030426,
      "num_tokens": 112907848.0,
      "step": 54405
    },
    {
      "entropy": 5.495687770843506,
      "epoch": 4.935595065312047,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002828028036135396,
      "loss": 4.9214,
      "mean_token_accuracy": 0.23013959378004073,
      "num_tokens": 112917344.0,
      "step": 54410
    },
    {
      "entropy": 5.480572032928467,
      "epoch": 4.936048621190131,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002827704665756101,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22563675940036773,
      "num_tokens": 112928004.0,
      "step": 54415
    },
    {
      "entropy": 5.473493242263794,
      "epoch": 4.936502177068215,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00028273812937698524,
      "loss": 4.856,
      "mean_token_accuracy": 0.22863715887069702,
      "num_tokens": 112938999.0,
      "step": 54420
    },
    {
      "entropy": 5.469941759109497,
      "epoch": 4.9369557329462985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028270579201833385,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22417243719100952,
      "num_tokens": 112948246.0,
      "step": 54425
    },
    {
      "entropy": 5.490860652923584,
      "epoch": 4.937409288824383,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002826734545003248,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22803116887807845,
      "num_tokens": 112958430.0,
      "step": 54430
    },
    {
      "entropy": 5.476547384262085,
      "epoch": 4.937862844702467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002826411168236268,
      "loss": 4.9008,
      "mean_token_accuracy": 0.22803355008363724,
      "num_tokens": 112969327.0,
      "step": 54435
    },
    {
      "entropy": 5.561001205444336,
      "epoch": 4.938316400580551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028260877898890843,
      "loss": 4.9625,
      "mean_token_accuracy": 0.22529940605163573,
      "num_tokens": 112980413.0,
      "step": 54440
    },
    {
      "entropy": 5.486183500289917,
      "epoch": 4.938769956458636,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00028257644099683864,
      "loss": 4.8618,
      "mean_token_accuracy": 0.228407983481884,
      "num_tokens": 112990924.0,
      "step": 54445
    },
    {
      "entropy": 5.496063709259033,
      "epoch": 4.93922351233672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002825441028480862,
      "loss": 4.957,
      "mean_token_accuracy": 0.20986407846212388,
      "num_tokens": 113001949.0,
      "step": 54450
    },
    {
      "entropy": 5.581991195678711,
      "epoch": 4.939677068214804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002825117645433196,
      "loss": 5.0357,
      "mean_token_accuracy": 0.21980217844247818,
      "num_tokens": 113012149.0,
      "step": 54455
    },
    {
      "entropy": 5.519558143615723,
      "epoch": 4.940130624092888,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028247942608320797,
      "loss": 4.9506,
      "mean_token_accuracy": 0.22329836189746857,
      "num_tokens": 113022783.0,
      "step": 54460
    },
    {
      "entropy": 5.4854024887084964,
      "epoch": 4.940584179970973,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028244708746841984,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22464623749256135,
      "num_tokens": 113032486.0,
      "step": 54465
    },
    {
      "entropy": 5.548636436462402,
      "epoch": 4.941037735849057,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002824147486996241,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2257835552096367,
      "num_tokens": 113042430.0,
      "step": 54470
    },
    {
      "entropy": 5.536951303482056,
      "epoch": 4.941491291727141,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028238240977748936,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21804095655679703,
      "num_tokens": 113052917.0,
      "step": 54475
    },
    {
      "entropy": 5.5276203632354735,
      "epoch": 4.941944847605225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028235007070268454,
      "loss": 4.955,
      "mean_token_accuracy": 0.22125871926546098,
      "num_tokens": 113063786.0,
      "step": 54480
    },
    {
      "entropy": 5.514996719360352,
      "epoch": 4.942398403483309,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028231773147587843,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22703657895326615,
      "num_tokens": 113073979.0,
      "step": 54485
    },
    {
      "entropy": 5.425254869461059,
      "epoch": 4.942851959361393,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002822853920977397,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22243181020021438,
      "num_tokens": 113085439.0,
      "step": 54490
    },
    {
      "entropy": 5.495853567123413,
      "epoch": 4.943305515239477,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002822530525689373,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21950509548187255,
      "num_tokens": 113096825.0,
      "step": 54495
    },
    {
      "entropy": 5.433024978637695,
      "epoch": 4.943759071117562,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028222071289013984,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23282776027917862,
      "num_tokens": 113107267.0,
      "step": 54500
    },
    {
      "entropy": 5.47907338142395,
      "epoch": 4.944212626995646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028218837306201623,
      "loss": 4.8358,
      "mean_token_accuracy": 0.2264660820364952,
      "num_tokens": 113118438.0,
      "step": 54505
    },
    {
      "entropy": 5.491107082366943,
      "epoch": 4.94466618287373,
      "grad_norm": 0.984375,
      "learning_rate": 0.00028215603308523517,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2257970914244652,
      "num_tokens": 113128361.0,
      "step": 54510
    },
    {
      "entropy": 5.510933923721313,
      "epoch": 4.945119738751814,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028212369296046566,
      "loss": 4.9834,
      "mean_token_accuracy": 0.22325991839170456,
      "num_tokens": 113138907.0,
      "step": 54515
    },
    {
      "entropy": 5.423193025588989,
      "epoch": 4.945573294629899,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028209135268837627,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2167103886604309,
      "num_tokens": 113149107.0,
      "step": 54520
    },
    {
      "entropy": 5.528092288970948,
      "epoch": 4.946026850507983,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028205901226963586,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2218063071370125,
      "num_tokens": 113159300.0,
      "step": 54525
    },
    {
      "entropy": 5.46118016242981,
      "epoch": 4.946480406386067,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028202667170491337,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23155918717384338,
      "num_tokens": 113168911.0,
      "step": 54530
    },
    {
      "entropy": 5.48734941482544,
      "epoch": 4.946933962264151,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002819943309948775,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22494154423475266,
      "num_tokens": 113179514.0,
      "step": 54535
    },
    {
      "entropy": 5.52489824295044,
      "epoch": 4.947387518142235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000281961990140197,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2245633825659752,
      "num_tokens": 113189732.0,
      "step": 54540
    },
    {
      "entropy": 5.525138425827026,
      "epoch": 4.947841074020319,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028192964914154086,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22513872236013413,
      "num_tokens": 113200012.0,
      "step": 54545
    },
    {
      "entropy": 5.492914438247681,
      "epoch": 4.948294629898403,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002818973079995778,
      "loss": 4.9604,
      "mean_token_accuracy": 0.22470972090959548,
      "num_tokens": 113210504.0,
      "step": 54550
    },
    {
      "entropy": 5.486193895339966,
      "epoch": 4.948748185776488,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002818649667149766,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22432801872491837,
      "num_tokens": 113220520.0,
      "step": 54555
    },
    {
      "entropy": 5.442224788665771,
      "epoch": 4.949201741654572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028183262528840616,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23315101563930513,
      "num_tokens": 113230355.0,
      "step": 54560
    },
    {
      "entropy": 5.49159984588623,
      "epoch": 4.949655297532656,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028180028372053537,
      "loss": 4.9199,
      "mean_token_accuracy": 0.21710179299116134,
      "num_tokens": 113240184.0,
      "step": 54565
    },
    {
      "entropy": 5.517309904098511,
      "epoch": 4.9501088534107405,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028176794201203293,
      "loss": 4.9664,
      "mean_token_accuracy": 0.22485513240098953,
      "num_tokens": 113249748.0,
      "step": 54570
    },
    {
      "entropy": 5.552989482879639,
      "epoch": 4.950562409288825,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028173560016356763,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2238339900970459,
      "num_tokens": 113260018.0,
      "step": 54575
    },
    {
      "entropy": 5.458161544799805,
      "epoch": 4.951015965166908,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002817032581758085,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2257182016968727,
      "num_tokens": 113269682.0,
      "step": 54580
    },
    {
      "entropy": 5.478200674057007,
      "epoch": 4.951469521044992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028167091604942423,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22388511002063752,
      "num_tokens": 113279485.0,
      "step": 54585
    },
    {
      "entropy": 5.562059783935547,
      "epoch": 4.951923076923077,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028163857378508386,
      "loss": 4.9824,
      "mean_token_accuracy": 0.22195178121328354,
      "num_tokens": 113290559.0,
      "step": 54590
    },
    {
      "entropy": 5.580182456970215,
      "epoch": 4.952376632801161,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028160623138345585,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2151132345199585,
      "num_tokens": 113300581.0,
      "step": 54595
    },
    {
      "entropy": 5.48119854927063,
      "epoch": 4.952830188679245,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028157388884520953,
      "loss": 4.8848,
      "mean_token_accuracy": 0.22324853986501694,
      "num_tokens": 113311261.0,
      "step": 54600
    },
    {
      "entropy": 5.486592960357666,
      "epoch": 4.9532837445573294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00028154154617101337,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22819592654705048,
      "num_tokens": 113321701.0,
      "step": 54605
    },
    {
      "entropy": 5.458574342727661,
      "epoch": 4.953737300435414,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028150920336153637,
      "loss": 4.893,
      "mean_token_accuracy": 0.22803890854120254,
      "num_tokens": 113332364.0,
      "step": 54610
    },
    {
      "entropy": 5.443449258804321,
      "epoch": 4.954190856313498,
      "grad_norm": 1.125,
      "learning_rate": 0.0002814768604174475,
      "loss": 4.8707,
      "mean_token_accuracy": 0.224417544901371,
      "num_tokens": 113342920.0,
      "step": 54615
    },
    {
      "entropy": 5.542629194259644,
      "epoch": 4.954644412191582,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002814445173394154,
      "loss": 4.9137,
      "mean_token_accuracy": 0.22140900790691376,
      "num_tokens": 113353602.0,
      "step": 54620
    },
    {
      "entropy": 5.444171667098999,
      "epoch": 4.9550979680696665,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028141217412810905,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22906980514526368,
      "num_tokens": 113362413.0,
      "step": 54625
    },
    {
      "entropy": 5.465959596633911,
      "epoch": 4.955551523947751,
      "grad_norm": 1.125,
      "learning_rate": 0.0002813798307841973,
      "loss": 4.88,
      "mean_token_accuracy": 0.2322030782699585,
      "num_tokens": 113373016.0,
      "step": 54630
    },
    {
      "entropy": 5.476529884338379,
      "epoch": 4.956005079825834,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002813474873083491,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22112055122852325,
      "num_tokens": 113384798.0,
      "step": 54635
    },
    {
      "entropy": 5.545580720901489,
      "epoch": 4.956458635703918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002813151437012331,
      "loss": 4.949,
      "mean_token_accuracy": 0.22374968677759172,
      "num_tokens": 113396266.0,
      "step": 54640
    },
    {
      "entropy": 5.515642786026001,
      "epoch": 4.956912191582003,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002812827999635185,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21919136494398117,
      "num_tokens": 113406475.0,
      "step": 54645
    },
    {
      "entropy": 5.505351209640503,
      "epoch": 4.957365747460087,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028125045609587383,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22857291549444197,
      "num_tokens": 113415835.0,
      "step": 54650
    },
    {
      "entropy": 5.512001371383667,
      "epoch": 4.957819303338171,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002812181120989683,
      "loss": 4.8933,
      "mean_token_accuracy": 0.223523311316967,
      "num_tokens": 113426359.0,
      "step": 54655
    },
    {
      "entropy": 5.510897970199585,
      "epoch": 4.9582728592162555,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002811857679734705,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21828763782978058,
      "num_tokens": 113437467.0,
      "step": 54660
    },
    {
      "entropy": 5.519833421707153,
      "epoch": 4.95872641509434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002811534237200494,
      "loss": 4.9865,
      "mean_token_accuracy": 0.22268702536821366,
      "num_tokens": 113447510.0,
      "step": 54665
    },
    {
      "entropy": 5.525196504592896,
      "epoch": 4.959179970972424,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002811210793393741,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2194933384656906,
      "num_tokens": 113457559.0,
      "step": 54670
    },
    {
      "entropy": 5.555708980560302,
      "epoch": 4.959633526850508,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028108873483211316,
      "loss": 5.0108,
      "mean_token_accuracy": 0.22091952264308928,
      "num_tokens": 113468589.0,
      "step": 54675
    },
    {
      "entropy": 5.473144006729126,
      "epoch": 4.960087082728592,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028105639019893573,
      "loss": 4.8909,
      "mean_token_accuracy": 0.22992948591709136,
      "num_tokens": 113478585.0,
      "step": 54680
    },
    {
      "entropy": 5.505551290512085,
      "epoch": 4.960540638606677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028102404544051056,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22883114963769913,
      "num_tokens": 113487980.0,
      "step": 54685
    },
    {
      "entropy": 5.484033536911011,
      "epoch": 4.96099419448476,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028099170055750667,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2306891143321991,
      "num_tokens": 113498123.0,
      "step": 54690
    },
    {
      "entropy": 5.509336948394775,
      "epoch": 4.9614477503628445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028095935555059277,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22266829162836074,
      "num_tokens": 113508685.0,
      "step": 54695
    },
    {
      "entropy": 5.470929765701294,
      "epoch": 4.961901306240929,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028092701042043794,
      "loss": 4.9485,
      "mean_token_accuracy": 0.22355875968933106,
      "num_tokens": 113519704.0,
      "step": 54700
    },
    {
      "entropy": 5.543742036819458,
      "epoch": 4.962354862119013,
      "grad_norm": 0.984375,
      "learning_rate": 0.000280894665167711,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21773332208395005,
      "num_tokens": 113529770.0,
      "step": 54705
    },
    {
      "entropy": 5.439230728149414,
      "epoch": 4.962808417997097,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002808623197930809,
      "loss": 4.7761,
      "mean_token_accuracy": 0.23720297664403917,
      "num_tokens": 113538667.0,
      "step": 54710
    },
    {
      "entropy": 5.540532636642456,
      "epoch": 4.9632619738751815,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002808299742972165,
      "loss": 4.932,
      "mean_token_accuracy": 0.22256590127944947,
      "num_tokens": 113549420.0,
      "step": 54715
    },
    {
      "entropy": 5.507911729812622,
      "epoch": 4.963715529753266,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028079762868078685,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22103143483400345,
      "num_tokens": 113559325.0,
      "step": 54720
    },
    {
      "entropy": 5.508272886276245,
      "epoch": 4.96416908563135,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002807652829444607,
      "loss": 4.8981,
      "mean_token_accuracy": 0.2258027747273445,
      "num_tokens": 113570152.0,
      "step": 54725
    },
    {
      "entropy": 5.446711730957031,
      "epoch": 4.964622641509434,
      "grad_norm": 0.96875,
      "learning_rate": 0.00028073293708890697,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2314901053905487,
      "num_tokens": 113580302.0,
      "step": 54730
    },
    {
      "entropy": 5.492828893661499,
      "epoch": 4.965076197387518,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028070059111479475,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21696748435497284,
      "num_tokens": 113590731.0,
      "step": 54735
    },
    {
      "entropy": 5.51592173576355,
      "epoch": 4.965529753265602,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002806682450227928,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2232059046626091,
      "num_tokens": 113600850.0,
      "step": 54740
    },
    {
      "entropy": 5.506237268447876,
      "epoch": 4.965983309143686,
      "grad_norm": 1.125,
      "learning_rate": 0.00028063589881357007,
      "loss": 4.936,
      "mean_token_accuracy": 0.221838016808033,
      "num_tokens": 113611559.0,
      "step": 54745
    },
    {
      "entropy": 5.437819719314575,
      "epoch": 4.9664368650217705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002806035524877955,
      "loss": 4.8076,
      "mean_token_accuracy": 0.22975514978170394,
      "num_tokens": 113621172.0,
      "step": 54750
    },
    {
      "entropy": 5.466757678985596,
      "epoch": 4.966890420899855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002805712060461381,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22331304401159285,
      "num_tokens": 113631573.0,
      "step": 54755
    },
    {
      "entropy": 5.5850917339324955,
      "epoch": 4.967343976777939,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028053885948926664,
      "loss": 5.0386,
      "mean_token_accuracy": 0.21566459983587266,
      "num_tokens": 113641990.0,
      "step": 54760
    },
    {
      "entropy": 5.5291175365448,
      "epoch": 4.967797532656023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002805065128178503,
      "loss": 4.93,
      "mean_token_accuracy": 0.22503428906202316,
      "num_tokens": 113651937.0,
      "step": 54765
    },
    {
      "entropy": 5.509579277038574,
      "epoch": 4.968251088534108,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028047416603255783,
      "loss": 4.9744,
      "mean_token_accuracy": 0.22462773472070693,
      "num_tokens": 113661150.0,
      "step": 54770
    },
    {
      "entropy": 5.485879898071289,
      "epoch": 4.968704644412192,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002804418191340582,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2234417602419853,
      "num_tokens": 113671636.0,
      "step": 54775
    },
    {
      "entropy": 5.535101270675659,
      "epoch": 4.969158200290276,
      "grad_norm": 1.125,
      "learning_rate": 0.00028040947212302033,
      "loss": 4.982,
      "mean_token_accuracy": 0.21772567182779312,
      "num_tokens": 113682051.0,
      "step": 54780
    },
    {
      "entropy": 5.486940574645996,
      "epoch": 4.96961175616836,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002803771250001132,
      "loss": 4.8219,
      "mean_token_accuracy": 0.2298119232058525,
      "num_tokens": 113691682.0,
      "step": 54785
    },
    {
      "entropy": 5.595443677902222,
      "epoch": 4.970065312046444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028034477776600576,
      "loss": 5.0294,
      "mean_token_accuracy": 0.21738992780447006,
      "num_tokens": 113702976.0,
      "step": 54790
    },
    {
      "entropy": 5.465690279006958,
      "epoch": 4.970518867924528,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028031243042136696,
      "loss": 4.8412,
      "mean_token_accuracy": 0.23280505388975142,
      "num_tokens": 113712871.0,
      "step": 54795
    },
    {
      "entropy": 5.463886165618897,
      "epoch": 4.970972423802612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002802800829668657,
      "loss": 4.8862,
      "mean_token_accuracy": 0.2179907664656639,
      "num_tokens": 113722425.0,
      "step": 54800
    },
    {
      "entropy": 5.504067659378052,
      "epoch": 4.9714259796806965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028024773540317105,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22290486991405487,
      "num_tokens": 113731851.0,
      "step": 54805
    },
    {
      "entropy": 5.481029224395752,
      "epoch": 4.971879535558781,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028021538773095184,
      "loss": 4.9573,
      "mean_token_accuracy": 0.21961503028869628,
      "num_tokens": 113742649.0,
      "step": 54810
    },
    {
      "entropy": 5.503659582138061,
      "epoch": 4.972333091436865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002801830399508771,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2252189576625824,
      "num_tokens": 113752620.0,
      "step": 54815
    },
    {
      "entropy": 5.424274682998657,
      "epoch": 4.972786647314949,
      "grad_norm": 1.125,
      "learning_rate": 0.00028015069206361584,
      "loss": 4.8743,
      "mean_token_accuracy": 0.23072244077920914,
      "num_tokens": 113762165.0,
      "step": 54820
    },
    {
      "entropy": 5.520236492156982,
      "epoch": 4.973240203193034,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028011834406983696,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22821362912654877,
      "num_tokens": 113773342.0,
      "step": 54825
    },
    {
      "entropy": 5.491857242584229,
      "epoch": 4.973693759071118,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002800859959702093,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22294944822788237,
      "num_tokens": 113783159.0,
      "step": 54830
    },
    {
      "entropy": 5.491335725784301,
      "epoch": 4.974147314949201,
      "grad_norm": 1.078125,
      "learning_rate": 0.000280053647765402,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22463326454162597,
      "num_tokens": 113794089.0,
      "step": 54835
    },
    {
      "entropy": 5.49736475944519,
      "epoch": 4.974600870827286,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028002129945608404,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21869278848171234,
      "num_tokens": 113804459.0,
      "step": 54840
    },
    {
      "entropy": 5.451222515106201,
      "epoch": 4.97505442670537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002799889510429242,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23479138761758805,
      "num_tokens": 113815032.0,
      "step": 54845
    },
    {
      "entropy": 5.481382656097412,
      "epoch": 4.975507982583454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002799566025265917,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22193410396575927,
      "num_tokens": 113825731.0,
      "step": 54850
    },
    {
      "entropy": 5.521695852279663,
      "epoch": 4.975961538461538,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002799242539077554,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2260073959827423,
      "num_tokens": 113835752.0,
      "step": 54855
    },
    {
      "entropy": 5.552551651000977,
      "epoch": 4.976415094339623,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002798919051870842,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22890870720148088,
      "num_tokens": 113845241.0,
      "step": 54860
    },
    {
      "entropy": 5.494087409973145,
      "epoch": 4.976868650217707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002798595563652472,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2178242787718773,
      "num_tokens": 113856066.0,
      "step": 54865
    },
    {
      "entropy": 5.434850072860717,
      "epoch": 4.977322206095791,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002798272074429133,
      "loss": 4.8376,
      "mean_token_accuracy": 0.22517310529947282,
      "num_tokens": 113866532.0,
      "step": 54870
    },
    {
      "entropy": 5.449157571792602,
      "epoch": 4.977775761973875,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002797948584207515,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22571404427289962,
      "num_tokens": 113876197.0,
      "step": 54875
    },
    {
      "entropy": 5.506764316558838,
      "epoch": 4.97822931785196,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002797625092994309,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22655236572027207,
      "num_tokens": 113886606.0,
      "step": 54880
    },
    {
      "entropy": 5.499774360656739,
      "epoch": 4.978682873730044,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027973016007962025,
      "loss": 4.902,
      "mean_token_accuracy": 0.2255020931363106,
      "num_tokens": 113897010.0,
      "step": 54885
    },
    {
      "entropy": 5.521834850311279,
      "epoch": 4.979136429608127,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002796978107619889,
      "loss": 4.8983,
      "mean_token_accuracy": 0.23272842019796372,
      "num_tokens": 113908762.0,
      "step": 54890
    },
    {
      "entropy": 5.511006784439087,
      "epoch": 4.979589985486212,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002796654613472054,
      "loss": 4.969,
      "mean_token_accuracy": 0.21444724947214128,
      "num_tokens": 113919164.0,
      "step": 54895
    },
    {
      "entropy": 5.536240768432617,
      "epoch": 4.980043541364296,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002796331118359391,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21652742922306062,
      "num_tokens": 113928975.0,
      "step": 54900
    },
    {
      "entropy": 5.469791460037231,
      "epoch": 4.98049709724238,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002796007622288588,
      "loss": 4.8487,
      "mean_token_accuracy": 0.23447393774986267,
      "num_tokens": 113937884.0,
      "step": 54905
    },
    {
      "entropy": 5.532653760910034,
      "epoch": 4.980950653120464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002795684125266336,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22438448667526245,
      "num_tokens": 113948750.0,
      "step": 54910
    },
    {
      "entropy": 5.563517379760742,
      "epoch": 4.981404208998549,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002795360627299324,
      "loss": 4.9297,
      "mean_token_accuracy": 0.22154538333415985,
      "num_tokens": 113960166.0,
      "step": 54915
    },
    {
      "entropy": 5.532948160171509,
      "epoch": 4.981857764876633,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00027950371283942443,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21958302408456803,
      "num_tokens": 113971279.0,
      "step": 54920
    },
    {
      "entropy": 5.511514806747437,
      "epoch": 4.982311320754717,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027947136285577846,
      "loss": 4.9523,
      "mean_token_accuracy": 0.22773263305425645,
      "num_tokens": 113980810.0,
      "step": 54925
    },
    {
      "entropy": 5.445932340621948,
      "epoch": 4.982764876632801,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027943901277966343,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22848851680755616,
      "num_tokens": 113990773.0,
      "step": 54930
    },
    {
      "entropy": 5.488036584854126,
      "epoch": 4.983218432510886,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002794066626117486,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2298087626695633,
      "num_tokens": 114001401.0,
      "step": 54935
    },
    {
      "entropy": 5.540445995330811,
      "epoch": 4.98367198838897,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002793743123527029,
      "loss": 4.9302,
      "mean_token_accuracy": 0.221894571185112,
      "num_tokens": 114012382.0,
      "step": 54940
    },
    {
      "entropy": 5.561772632598877,
      "epoch": 4.984125544267053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002793419620031952,
      "loss": 4.952,
      "mean_token_accuracy": 0.21940550208091736,
      "num_tokens": 114022395.0,
      "step": 54945
    },
    {
      "entropy": 5.522034502029419,
      "epoch": 4.984579100145138,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027930961156389473,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2171648249030113,
      "num_tokens": 114032629.0,
      "step": 54950
    },
    {
      "entropy": 5.4844642162323,
      "epoch": 4.985032656023222,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027927726103547034,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2297092616558075,
      "num_tokens": 114042642.0,
      "step": 54955
    },
    {
      "entropy": 5.484522676467895,
      "epoch": 4.985486211901306,
      "grad_norm": 0.953125,
      "learning_rate": 0.0002792449104185911,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21393949687480926,
      "num_tokens": 114052972.0,
      "step": 54960
    },
    {
      "entropy": 5.512157344818116,
      "epoch": 4.98593976777939,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027921255971392607,
      "loss": 5.0127,
      "mean_token_accuracy": 0.21576126664876938,
      "num_tokens": 114064422.0,
      "step": 54965
    },
    {
      "entropy": 5.560077381134033,
      "epoch": 4.986393323657475,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002791802089221442,
      "loss": 4.942,
      "mean_token_accuracy": 0.22095310389995576,
      "num_tokens": 114075611.0,
      "step": 54970
    },
    {
      "entropy": 5.580923652648925,
      "epoch": 4.986846879535559,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027914785804391455,
      "loss": 4.9575,
      "mean_token_accuracy": 0.22373263239860536,
      "num_tokens": 114086686.0,
      "step": 54975
    },
    {
      "entropy": 5.543174982070923,
      "epoch": 4.987300435413643,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027911550707990614,
      "loss": 4.862,
      "mean_token_accuracy": 0.2262525036931038,
      "num_tokens": 114097968.0,
      "step": 54980
    },
    {
      "entropy": 5.4530104637146,
      "epoch": 4.9877539912917275,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027908315603078796,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22414759695529937,
      "num_tokens": 114108323.0,
      "step": 54985
    },
    {
      "entropy": 5.405993509292602,
      "epoch": 4.988207547169811,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027905080489722914,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22755447924137115,
      "num_tokens": 114118129.0,
      "step": 54990
    },
    {
      "entropy": 5.504064989089966,
      "epoch": 4.988661103047895,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002790184536798985,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21737308353185653,
      "num_tokens": 114127426.0,
      "step": 54995
    },
    {
      "entropy": 5.495132493972778,
      "epoch": 4.989114658925979,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027898610237946537,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23167398273944856,
      "num_tokens": 114137216.0,
      "step": 55000
    },
    {
      "entropy": 5.532276439666748,
      "epoch": 4.989568214804064,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002789537509965986,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21823213696479798,
      "num_tokens": 114148617.0,
      "step": 55005
    },
    {
      "entropy": 5.57807936668396,
      "epoch": 4.990021770682148,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002789213995319672,
      "loss": 4.937,
      "mean_token_accuracy": 0.22083556950092315,
      "num_tokens": 114159512.0,
      "step": 55010
    },
    {
      "entropy": 5.457546281814575,
      "epoch": 4.990475326560232,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002788890479862403,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22667690366506577,
      "num_tokens": 114169249.0,
      "step": 55015
    },
    {
      "entropy": 5.522327470779419,
      "epoch": 4.9909288824383164,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027885669636008685,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21725826412439347,
      "num_tokens": 114179622.0,
      "step": 55020
    },
    {
      "entropy": 5.477546834945679,
      "epoch": 4.991382438316401,
      "grad_norm": 1.140625,
      "learning_rate": 0.000278824344654176,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23352111130952835,
      "num_tokens": 114189712.0,
      "step": 55025
    },
    {
      "entropy": 5.560367155075073,
      "epoch": 4.991835994194485,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027879199286917665,
      "loss": 5.0056,
      "mean_token_accuracy": 0.21404151022434234,
      "num_tokens": 114200074.0,
      "step": 55030
    },
    {
      "entropy": 5.488446950912476,
      "epoch": 4.992289550072569,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000278759641005758,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2243913233280182,
      "num_tokens": 114209659.0,
      "step": 55035
    },
    {
      "entropy": 5.500846290588379,
      "epoch": 4.9927431059506535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000278727289064589,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22481512427330017,
      "num_tokens": 114219765.0,
      "step": 55040
    },
    {
      "entropy": 5.507067441940308,
      "epoch": 4.993196661828737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002786949370463387,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22816997915506362,
      "num_tokens": 114229654.0,
      "step": 55045
    },
    {
      "entropy": 5.503411865234375,
      "epoch": 4.993650217706821,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002786625849516762,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21902555823326111,
      "num_tokens": 114240504.0,
      "step": 55050
    },
    {
      "entropy": 5.518369054794311,
      "epoch": 4.994103773584905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002786302327812704,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21188020408153535,
      "num_tokens": 114250748.0,
      "step": 55055
    },
    {
      "entropy": 5.538392448425293,
      "epoch": 4.99455732946299,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027859788053579055,
      "loss": 4.9658,
      "mean_token_accuracy": 0.21858011037111283,
      "num_tokens": 114261655.0,
      "step": 55060
    },
    {
      "entropy": 5.48295226097107,
      "epoch": 4.995010885341074,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027856552821590557,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22178115844726562,
      "num_tokens": 114271193.0,
      "step": 55065
    },
    {
      "entropy": 5.483733654022217,
      "epoch": 4.995464441219158,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002785331758222846,
      "loss": 4.8971,
      "mean_token_accuracy": 0.224680033326149,
      "num_tokens": 114281461.0,
      "step": 55070
    },
    {
      "entropy": 5.494833946228027,
      "epoch": 4.9959179970972425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002785008233555967,
      "loss": 4.9534,
      "mean_token_accuracy": 0.21581086665391921,
      "num_tokens": 114290865.0,
      "step": 55075
    },
    {
      "entropy": 5.553004407882691,
      "epoch": 4.996371552975327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002784684708165107,
      "loss": 4.959,
      "mean_token_accuracy": 0.22085859328508378,
      "num_tokens": 114301811.0,
      "step": 55080
    },
    {
      "entropy": 5.560033893585205,
      "epoch": 4.996825108853411,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027843611820569595,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21856225579977034,
      "num_tokens": 114313070.0,
      "step": 55085
    },
    {
      "entropy": 5.529986906051636,
      "epoch": 4.997278664731495,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027840376552382145,
      "loss": 4.9895,
      "mean_token_accuracy": 0.21597571074962615,
      "num_tokens": 114323670.0,
      "step": 55090
    },
    {
      "entropy": 5.422468042373657,
      "epoch": 4.99773222060958,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027837141277155616,
      "loss": 4.8219,
      "mean_token_accuracy": 0.2286313772201538,
      "num_tokens": 114334367.0,
      "step": 55095
    },
    {
      "entropy": 5.423502731323242,
      "epoch": 4.998185776487663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027833905994956913,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2347673937678337,
      "num_tokens": 114344759.0,
      "step": 55100
    },
    {
      "entropy": 5.483346843719483,
      "epoch": 4.998639332365747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002783067070585296,
      "loss": 4.8717,
      "mean_token_accuracy": 0.21633470505476,
      "num_tokens": 114355200.0,
      "step": 55105
    },
    {
      "entropy": 5.4939642429351805,
      "epoch": 4.9990928882438315,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00027827435409910644,
      "loss": 4.8804,
      "mean_token_accuracy": 0.23216792941093445,
      "num_tokens": 114365333.0,
      "step": 55110
    },
    {
      "entropy": 5.498613357543945,
      "epoch": 4.999546444121916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002782420010719689,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22562285810708999,
      "num_tokens": 114375965.0,
      "step": 55115
    },
    {
      "entropy": 5.5192924499511715,
      "epoch": 5.0,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027820964797778587,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21762098222970963,
      "num_tokens": 114386175.0,
      "step": 55120
    },
    {
      "entropy": 5.442185592651367,
      "epoch": 5.000453555878084,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027817729481722657,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2338218480348587,
      "num_tokens": 114395139.0,
      "step": 55125
    },
    {
      "entropy": 5.453617715835572,
      "epoch": 5.0009071117561685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027814494159095996,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23846292197704316,
      "num_tokens": 114405378.0,
      "step": 55130
    },
    {
      "entropy": 5.486904668807983,
      "epoch": 5.001360667634253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027811258829965515,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2281859502196312,
      "num_tokens": 114415920.0,
      "step": 55135
    },
    {
      "entropy": 5.529765033721924,
      "epoch": 5.001814223512337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002780802349439812,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21713445335626602,
      "num_tokens": 114426177.0,
      "step": 55140
    },
    {
      "entropy": 5.5101720809936525,
      "epoch": 5.002267779390421,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027804788152460723,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22616015672683715,
      "num_tokens": 114436221.0,
      "step": 55145
    },
    {
      "entropy": 5.381057929992676,
      "epoch": 5.002721335268505,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002780155280422023,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2396004915237427,
      "num_tokens": 114445828.0,
      "step": 55150
    },
    {
      "entropy": 5.485676050186157,
      "epoch": 5.003174891146589,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027798317449743553,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22584061622619628,
      "num_tokens": 114454792.0,
      "step": 55155
    },
    {
      "entropy": 5.445809555053711,
      "epoch": 5.003628447024673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002779508208909759,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22829709947109222,
      "num_tokens": 114465357.0,
      "step": 55160
    },
    {
      "entropy": 5.479586029052735,
      "epoch": 5.0040820029027575,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002779184672234925,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23128732591867446,
      "num_tokens": 114475947.0,
      "step": 55165
    },
    {
      "entropy": 5.4725195407867435,
      "epoch": 5.004535558780842,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002778861134956545,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22680184692144395,
      "num_tokens": 114486335.0,
      "step": 55170
    },
    {
      "entropy": 5.5309065818786625,
      "epoch": 5.004989114658926,
      "grad_norm": 1.0625,
      "learning_rate": 0.000277853759708131,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22693614065647125,
      "num_tokens": 114495959.0,
      "step": 55175
    },
    {
      "entropy": 5.527949571609497,
      "epoch": 5.00544267053701,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002778214058615909,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2227547973394394,
      "num_tokens": 114506328.0,
      "step": 55180
    },
    {
      "entropy": 5.445960187911988,
      "epoch": 5.005896226415095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002777890519567035,
      "loss": 4.8296,
      "mean_token_accuracy": 0.2348012864589691,
      "num_tokens": 114515368.0,
      "step": 55185
    },
    {
      "entropy": 5.414036464691162,
      "epoch": 5.006349782293179,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027775669799413774,
      "loss": 4.8815,
      "mean_token_accuracy": 0.23066049069166183,
      "num_tokens": 114525894.0,
      "step": 55190
    },
    {
      "entropy": 5.494455289840698,
      "epoch": 5.006803338171263,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027772434397456284,
      "loss": 4.9659,
      "mean_token_accuracy": 0.2250737115740776,
      "num_tokens": 114535739.0,
      "step": 55195
    },
    {
      "entropy": 5.454811763763428,
      "epoch": 5.0072568940493465,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027769198989864764,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23722704946994783,
      "num_tokens": 114546519.0,
      "step": 55200
    },
    {
      "entropy": 5.47035231590271,
      "epoch": 5.007710449927431,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0002776596357670616,
      "loss": 4.8222,
      "mean_token_accuracy": 0.2279689148068428,
      "num_tokens": 114556523.0,
      "step": 55205
    },
    {
      "entropy": 5.531890773773194,
      "epoch": 5.008164005805515,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027762728158047353,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2241772085428238,
      "num_tokens": 114566255.0,
      "step": 55210
    },
    {
      "entropy": 5.492157793045044,
      "epoch": 5.008617561683599,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00027759492733955257,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23000467121601104,
      "num_tokens": 114576686.0,
      "step": 55215
    },
    {
      "entropy": 5.472742652893066,
      "epoch": 5.0090711175616836,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002775625730449679,
      "loss": 4.8721,
      "mean_token_accuracy": 0.22387513816356658,
      "num_tokens": 114586606.0,
      "step": 55220
    },
    {
      "entropy": 5.471456241607666,
      "epoch": 5.009524673439768,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00027753021869738855,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2283127024769783,
      "num_tokens": 114598450.0,
      "step": 55225
    },
    {
      "entropy": 5.570858573913574,
      "epoch": 5.009978229317852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027749786429748366,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22313912957906723,
      "num_tokens": 114608761.0,
      "step": 55230
    },
    {
      "entropy": 5.525013732910156,
      "epoch": 5.010431785195936,
      "grad_norm": 1.0,
      "learning_rate": 0.0002774655098459223,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22562316805124283,
      "num_tokens": 114619100.0,
      "step": 55235
    },
    {
      "entropy": 5.547834587097168,
      "epoch": 5.010885341074021,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002774331553433735,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22142431288957595,
      "num_tokens": 114629594.0,
      "step": 55240
    },
    {
      "entropy": 5.436564826965332,
      "epoch": 5.011338896952105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002774008007905065,
      "loss": 4.8021,
      "mean_token_accuracy": 0.22247554063796998,
      "num_tokens": 114639090.0,
      "step": 55245
    },
    {
      "entropy": 5.4648566246032715,
      "epoch": 5.011792452830188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027736844618799026,
      "loss": 4.829,
      "mean_token_accuracy": 0.23262514024972916,
      "num_tokens": 114648707.0,
      "step": 55250
    },
    {
      "entropy": 5.4480791091918945,
      "epoch": 5.0122460087082725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002773360915364941,
      "loss": 4.9068,
      "mean_token_accuracy": 0.22193971872329712,
      "num_tokens": 114659534.0,
      "step": 55255
    },
    {
      "entropy": 5.476681470870972,
      "epoch": 5.012699564586357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002773037368366868,
      "loss": 4.825,
      "mean_token_accuracy": 0.2303838849067688,
      "num_tokens": 114670094.0,
      "step": 55260
    },
    {
      "entropy": 5.510905075073242,
      "epoch": 5.013153120464441,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002772713820892377,
      "loss": 4.8941,
      "mean_token_accuracy": 0.23212112188339235,
      "num_tokens": 114680807.0,
      "step": 55265
    },
    {
      "entropy": 5.509872722625732,
      "epoch": 5.013606676342525,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002772390272948159,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22926875948905945,
      "num_tokens": 114691334.0,
      "step": 55270
    },
    {
      "entropy": 5.4890156269073485,
      "epoch": 5.01406023222061,
      "grad_norm": 1.0,
      "learning_rate": 0.00027720667245409035,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21947411745786666,
      "num_tokens": 114702519.0,
      "step": 55275
    },
    {
      "entropy": 5.505037641525268,
      "epoch": 5.014513788098694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027717431756773024,
      "loss": 4.8485,
      "mean_token_accuracy": 0.23400871753692626,
      "num_tokens": 114712893.0,
      "step": 55280
    },
    {
      "entropy": 5.5332763671875,
      "epoch": 5.014967343976778,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002771419626364048,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2152680918574333,
      "num_tokens": 114722101.0,
      "step": 55285
    },
    {
      "entropy": 5.502281522750854,
      "epoch": 5.015420899854862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027710960766078296,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22522107660770416,
      "num_tokens": 114733184.0,
      "step": 55290
    },
    {
      "entropy": 5.440610885620117,
      "epoch": 5.015874455732947,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00027707725264153386,
      "loss": 4.7517,
      "mean_token_accuracy": 0.23199809044599534,
      "num_tokens": 114743553.0,
      "step": 55295
    },
    {
      "entropy": 5.4918567657470705,
      "epoch": 5.01632801161103,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027704489757932667,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22993380576372147,
      "num_tokens": 114754655.0,
      "step": 55300
    },
    {
      "entropy": 5.497515106201172,
      "epoch": 5.016781567489114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002770125424748305,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2160625472664833,
      "num_tokens": 114764918.0,
      "step": 55305
    },
    {
      "entropy": 5.511150979995728,
      "epoch": 5.017235123367199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002769801873287145,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22422921508550644,
      "num_tokens": 114775968.0,
      "step": 55310
    },
    {
      "entropy": 5.504076051712036,
      "epoch": 5.017688679245283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027694783214164757,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22932385504245759,
      "num_tokens": 114786363.0,
      "step": 55315
    },
    {
      "entropy": 5.508125829696655,
      "epoch": 5.018142235123367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002769154769142991,
      "loss": 4.883,
      "mean_token_accuracy": 0.22634349167346954,
      "num_tokens": 114797034.0,
      "step": 55320
    },
    {
      "entropy": 5.481206607818604,
      "epoch": 5.018595791001451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002768831216473381,
      "loss": 4.8151,
      "mean_token_accuracy": 0.227792589366436,
      "num_tokens": 114806493.0,
      "step": 55325
    },
    {
      "entropy": 5.4931478023529055,
      "epoch": 5.019049346879536,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00027685076634143354,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21835681349039077,
      "num_tokens": 114816986.0,
      "step": 55330
    },
    {
      "entropy": 5.4654745101928714,
      "epoch": 5.01950290275762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002768184109972547,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2224149540066719,
      "num_tokens": 114827592.0,
      "step": 55335
    },
    {
      "entropy": 5.528166818618774,
      "epoch": 5.019956458635704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027678605561547073,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22652609199285506,
      "num_tokens": 114839080.0,
      "step": 55340
    },
    {
      "entropy": 5.55200719833374,
      "epoch": 5.020410014513788,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002767537001967506,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22394368350505828,
      "num_tokens": 114848816.0,
      "step": 55345
    },
    {
      "entropy": 5.498950672149658,
      "epoch": 5.020863570391873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027672134474176347,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2315750539302826,
      "num_tokens": 114858781.0,
      "step": 55350
    },
    {
      "entropy": 5.438818359375,
      "epoch": 5.021317126269956,
      "grad_norm": 1.0,
      "learning_rate": 0.00027668898925117856,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22858970761299133,
      "num_tokens": 114869257.0,
      "step": 55355
    },
    {
      "entropy": 5.4539008140563965,
      "epoch": 5.02177068214804,
      "grad_norm": 1.125,
      "learning_rate": 0.0002766566337256649,
      "loss": 4.8281,
      "mean_token_accuracy": 0.24423001557588578,
      "num_tokens": 114879878.0,
      "step": 55360
    },
    {
      "entropy": 5.542954778671264,
      "epoch": 5.022224238026125,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027662427816589163,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2241322860121727,
      "num_tokens": 114890253.0,
      "step": 55365
    },
    {
      "entropy": 5.527041339874268,
      "epoch": 5.022677793904209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002765919225725279,
      "loss": 4.886,
      "mean_token_accuracy": 0.22932648062705993,
      "num_tokens": 114900715.0,
      "step": 55370
    },
    {
      "entropy": 5.464224720001221,
      "epoch": 5.023131349782293,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002765595669462428,
      "loss": 4.8203,
      "mean_token_accuracy": 0.22979944497346877,
      "num_tokens": 114910663.0,
      "step": 55375
    },
    {
      "entropy": 5.523819255828857,
      "epoch": 5.023584905660377,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002765272112877054,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21797103434801102,
      "num_tokens": 114920844.0,
      "step": 55380
    },
    {
      "entropy": 5.514798974990844,
      "epoch": 5.024038461538462,
      "grad_norm": 1.0,
      "learning_rate": 0.000276494855597585,
      "loss": 4.8609,
      "mean_token_accuracy": 0.23077383190393447,
      "num_tokens": 114931701.0,
      "step": 55385
    },
    {
      "entropy": 5.539121341705322,
      "epoch": 5.024492017416546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027646249987655047,
      "loss": 4.8826,
      "mean_token_accuracy": 0.23188445270061492,
      "num_tokens": 114941764.0,
      "step": 55390
    },
    {
      "entropy": 5.489105176925659,
      "epoch": 5.02494557329463,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002764301441252712,
      "loss": 4.807,
      "mean_token_accuracy": 0.2329765647649765,
      "num_tokens": 114952126.0,
      "step": 55395
    },
    {
      "entropy": 5.4245484352111815,
      "epoch": 5.0253991291727145,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027639778834441606,
      "loss": 4.833,
      "mean_token_accuracy": 0.23572560548782348,
      "num_tokens": 114961952.0,
      "step": 55400
    },
    {
      "entropy": 5.5291646957397464,
      "epoch": 5.025852685050798,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002763654325346544,
      "loss": 4.9372,
      "mean_token_accuracy": 0.22568818777799607,
      "num_tokens": 114971338.0,
      "step": 55405
    },
    {
      "entropy": 5.5362548828125,
      "epoch": 5.026306240928882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027633307669665524,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22980718910694123,
      "num_tokens": 114981792.0,
      "step": 55410
    },
    {
      "entropy": 5.480351686477661,
      "epoch": 5.026759796806966,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002763007208310876,
      "loss": 4.7995,
      "mean_token_accuracy": 0.23284534960985184,
      "num_tokens": 114992619.0,
      "step": 55415
    },
    {
      "entropy": 5.481419849395752,
      "epoch": 5.027213352685051,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027626836493862086,
      "loss": 4.8273,
      "mean_token_accuracy": 0.23283834904432296,
      "num_tokens": 115003551.0,
      "step": 55420
    },
    {
      "entropy": 5.556876802444458,
      "epoch": 5.027666908563135,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00027623600901992393,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2210737034678459,
      "num_tokens": 115013155.0,
      "step": 55425
    },
    {
      "entropy": 5.477897262573242,
      "epoch": 5.028120464441219,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027620365307566607,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22144166380167007,
      "num_tokens": 115023673.0,
      "step": 55430
    },
    {
      "entropy": 5.497369623184204,
      "epoch": 5.0285740203193035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002761712971065163,
      "loss": 4.9137,
      "mean_token_accuracy": 0.226165272295475,
      "num_tokens": 115033597.0,
      "step": 55435
    },
    {
      "entropy": 5.488608884811401,
      "epoch": 5.029027576197388,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027613894111314384,
      "loss": 4.804,
      "mean_token_accuracy": 0.23355222195386888,
      "num_tokens": 115043732.0,
      "step": 55440
    },
    {
      "entropy": 5.49675669670105,
      "epoch": 5.029481132075472,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027610658509621776,
      "loss": 4.8676,
      "mean_token_accuracy": 0.23110130280256272,
      "num_tokens": 115054124.0,
      "step": 55445
    },
    {
      "entropy": 5.454931974411011,
      "epoch": 5.029934687953556,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002760742290564072,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24232590943574905,
      "num_tokens": 115064513.0,
      "step": 55450
    },
    {
      "entropy": 5.509035205841064,
      "epoch": 5.03038824383164,
      "grad_norm": 1.125,
      "learning_rate": 0.00027604187299438146,
      "loss": 4.9386,
      "mean_token_accuracy": 0.23000385612249374,
      "num_tokens": 115074060.0,
      "step": 55455
    },
    {
      "entropy": 5.547656488418579,
      "epoch": 5.030841799709724,
      "grad_norm": 1.125,
      "learning_rate": 0.00027600951691080937,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21724499613046647,
      "num_tokens": 115084615.0,
      "step": 55460
    },
    {
      "entropy": 5.477855205535889,
      "epoch": 5.031295355587808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027597716080636026,
      "loss": 4.758,
      "mean_token_accuracy": 0.24062436819076538,
      "num_tokens": 115095899.0,
      "step": 55465
    },
    {
      "entropy": 5.4190356731414795,
      "epoch": 5.031748911465892,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002759448046817033,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22470982372760773,
      "num_tokens": 115107121.0,
      "step": 55470
    },
    {
      "entropy": 5.5201374053955075,
      "epoch": 5.032202467343977,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002759124485375074,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22716695964336395,
      "num_tokens": 115117842.0,
      "step": 55475
    },
    {
      "entropy": 5.446328353881836,
      "epoch": 5.032656023222061,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002758800923744419,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23906033784151076,
      "num_tokens": 115127268.0,
      "step": 55480
    },
    {
      "entropy": 5.461370897293091,
      "epoch": 5.033109579100145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027584773619317583,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23538366556167603,
      "num_tokens": 115137102.0,
      "step": 55485
    },
    {
      "entropy": 5.535008382797241,
      "epoch": 5.0335631349782295,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002758153799943784,
      "loss": 4.9268,
      "mean_token_accuracy": 0.22501951903104783,
      "num_tokens": 115147822.0,
      "step": 55490
    },
    {
      "entropy": 5.693020534515381,
      "epoch": 5.034016690856314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027578302377871867,
      "loss": 5.0783,
      "mean_token_accuracy": 0.21051571816205977,
      "num_tokens": 115158163.0,
      "step": 55495
    },
    {
      "entropy": 5.449180364608765,
      "epoch": 5.034470246734398,
      "grad_norm": 0.984375,
      "learning_rate": 0.00027575066754686583,
      "loss": 4.8497,
      "mean_token_accuracy": 0.21729307770729064,
      "num_tokens": 115169224.0,
      "step": 55500
    },
    {
      "entropy": 5.43520073890686,
      "epoch": 5.034923802612482,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027571831129948895,
      "loss": 4.788,
      "mean_token_accuracy": 0.23335777670145036,
      "num_tokens": 115179357.0,
      "step": 55505
    },
    {
      "entropy": 5.537153863906861,
      "epoch": 5.035377358490566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002756859550372573,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2265951544046402,
      "num_tokens": 115190245.0,
      "step": 55510
    },
    {
      "entropy": 5.565239381790161,
      "epoch": 5.03583091436865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002756535987608398,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22332887947559357,
      "num_tokens": 115201106.0,
      "step": 55515
    },
    {
      "entropy": 5.534576177597046,
      "epoch": 5.036284470246734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002756212424709058,
      "loss": 4.8626,
      "mean_token_accuracy": 0.23123902678489686,
      "num_tokens": 115211822.0,
      "step": 55520
    },
    {
      "entropy": 5.48705267906189,
      "epoch": 5.0367380261248185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027558888616812435,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22780181020498275,
      "num_tokens": 115222662.0,
      "step": 55525
    },
    {
      "entropy": 5.5093231201171875,
      "epoch": 5.037191582002903,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002755565298531645,
      "loss": 4.907,
      "mean_token_accuracy": 0.22506503164768218,
      "num_tokens": 115232534.0,
      "step": 55530
    },
    {
      "entropy": 5.53064923286438,
      "epoch": 5.037645137880987,
      "grad_norm": 1.0,
      "learning_rate": 0.0002755241735266956,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22418875843286515,
      "num_tokens": 115243669.0,
      "step": 55535
    },
    {
      "entropy": 5.497088384628296,
      "epoch": 5.038098693759071,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002754918171893865,
      "loss": 4.8222,
      "mean_token_accuracy": 0.24024057686328887,
      "num_tokens": 115253482.0,
      "step": 55540
    },
    {
      "entropy": 5.474453735351562,
      "epoch": 5.0385522496371555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027545946084190666,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22238099724054336,
      "num_tokens": 115263037.0,
      "step": 55545
    },
    {
      "entropy": 5.5247986793518065,
      "epoch": 5.03900580551524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002754271044849249,
      "loss": 4.9126,
      "mean_token_accuracy": 0.22385069727897644,
      "num_tokens": 115273135.0,
      "step": 55550
    },
    {
      "entropy": 5.583429336547852,
      "epoch": 5.039459361393324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027539474811911065,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22580430507659913,
      "num_tokens": 115283482.0,
      "step": 55555
    },
    {
      "entropy": 5.434324312210083,
      "epoch": 5.0399129172714074,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002753623917451329,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2349749431014061,
      "num_tokens": 115293855.0,
      "step": 55560
    },
    {
      "entropy": 5.4826044082641605,
      "epoch": 5.040366473149492,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002753300353636607,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2266663447022438,
      "num_tokens": 115304162.0,
      "step": 55565
    },
    {
      "entropy": 5.564769840240478,
      "epoch": 5.040820029027576,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027529767897536327,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22484886795282363,
      "num_tokens": 115315108.0,
      "step": 55570
    },
    {
      "entropy": 5.486380672454834,
      "epoch": 5.04127358490566,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00027526532258090986,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2201646089553833,
      "num_tokens": 115325434.0,
      "step": 55575
    },
    {
      "entropy": 5.521027946472168,
      "epoch": 5.0417271407837445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027523296618096947,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22392463088035583,
      "num_tokens": 115335410.0,
      "step": 55580
    },
    {
      "entropy": 5.527436828613281,
      "epoch": 5.042180696661829,
      "grad_norm": 1.125,
      "learning_rate": 0.00027520060977621126,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2352551966905594,
      "num_tokens": 115345838.0,
      "step": 55585
    },
    {
      "entropy": 5.433565425872803,
      "epoch": 5.042634252539913,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002751682533673044,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2355193704366684,
      "num_tokens": 115355988.0,
      "step": 55590
    },
    {
      "entropy": 5.411842107772827,
      "epoch": 5.043087808417997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002751358969549181,
      "loss": 4.8351,
      "mean_token_accuracy": 0.22606998234987258,
      "num_tokens": 115366942.0,
      "step": 55595
    },
    {
      "entropy": 5.467719507217407,
      "epoch": 5.043541364296082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027510354053972123,
      "loss": 4.7755,
      "mean_token_accuracy": 0.24555230736732483,
      "num_tokens": 115376948.0,
      "step": 55600
    },
    {
      "entropy": 5.506367206573486,
      "epoch": 5.043994920174166,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027507118412238324,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2346438229084015,
      "num_tokens": 115387831.0,
      "step": 55605
    },
    {
      "entropy": 5.488424015045166,
      "epoch": 5.044448476052249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002750388277035732,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22684485465288162,
      "num_tokens": 115397511.0,
      "step": 55610
    },
    {
      "entropy": 5.428844022750854,
      "epoch": 5.0449020319303335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002750064712839601,
      "loss": 4.8265,
      "mean_token_accuracy": 0.23097184896469117,
      "num_tokens": 115408049.0,
      "step": 55615
    },
    {
      "entropy": 5.49310302734375,
      "epoch": 5.045355587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027497411486421323,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22868202924728392,
      "num_tokens": 115418811.0,
      "step": 55620
    },
    {
      "entropy": 5.5224545955657955,
      "epoch": 5.045809143686502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027494175844500164,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23177893906831742,
      "num_tokens": 115428324.0,
      "step": 55625
    },
    {
      "entropy": 5.476614618301392,
      "epoch": 5.046262699564586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002749094020269945,
      "loss": 4.833,
      "mean_token_accuracy": 0.228506638109684,
      "num_tokens": 115437891.0,
      "step": 55630
    },
    {
      "entropy": 5.44257001876831,
      "epoch": 5.0467162554426706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000274877045610861,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22578754872083664,
      "num_tokens": 115448100.0,
      "step": 55635
    },
    {
      "entropy": 5.447836446762085,
      "epoch": 5.047169811320755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002748446891972701,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23760270327329636,
      "num_tokens": 115458530.0,
      "step": 55640
    },
    {
      "entropy": 5.476264524459839,
      "epoch": 5.047623367198839,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027481233278689124,
      "loss": 4.8475,
      "mean_token_accuracy": 0.23155716359615325,
      "num_tokens": 115469222.0,
      "step": 55645
    },
    {
      "entropy": 5.521099090576172,
      "epoch": 5.048076923076923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002747799763803932,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22514349669218064,
      "num_tokens": 115478704.0,
      "step": 55650
    },
    {
      "entropy": 5.490459251403808,
      "epoch": 5.048530478955008,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027474761997844536,
      "loss": 4.8948,
      "mean_token_accuracy": 0.22290227115154265,
      "num_tokens": 115488430.0,
      "step": 55655
    },
    {
      "entropy": 5.479153060913086,
      "epoch": 5.048984034833092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002747152635817169,
      "loss": 4.916,
      "mean_token_accuracy": 0.22642832249403,
      "num_tokens": 115499010.0,
      "step": 55660
    },
    {
      "entropy": 5.468541383743286,
      "epoch": 5.049437590711175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002746829071908768,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23094337582588195,
      "num_tokens": 115508278.0,
      "step": 55665
    },
    {
      "entropy": 5.4869366645812985,
      "epoch": 5.0498911465892595,
      "grad_norm": 1.125,
      "learning_rate": 0.0002746505508065942,
      "loss": 4.7907,
      "mean_token_accuracy": 0.22884036600589752,
      "num_tokens": 115517977.0,
      "step": 55670
    },
    {
      "entropy": 5.497477674484253,
      "epoch": 5.050344702467344,
      "grad_norm": 1.0,
      "learning_rate": 0.00027461819442953846,
      "loss": 4.8208,
      "mean_token_accuracy": 0.22783585786819457,
      "num_tokens": 115528533.0,
      "step": 55675
    },
    {
      "entropy": 5.4629661560058596,
      "epoch": 5.050798258345428,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002745858380603784,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2326045736670494,
      "num_tokens": 115537930.0,
      "step": 55680
    },
    {
      "entropy": 5.5041797161102295,
      "epoch": 5.051251814223512,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002745534816997834,
      "loss": 4.935,
      "mean_token_accuracy": 0.2165326490998268,
      "num_tokens": 115547436.0,
      "step": 55685
    },
    {
      "entropy": 5.475390100479126,
      "epoch": 5.051705370101597,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027452112534842246,
      "loss": 4.82,
      "mean_token_accuracy": 0.23002816438674928,
      "num_tokens": 115559313.0,
      "step": 55690
    },
    {
      "entropy": 5.552815103530884,
      "epoch": 5.052158925979681,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002744887690069648,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22628372758626938,
      "num_tokens": 115568773.0,
      "step": 55695
    },
    {
      "entropy": 5.484169626235962,
      "epoch": 5.052612481857765,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027445641267607947,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22528289556503295,
      "num_tokens": 115579402.0,
      "step": 55700
    },
    {
      "entropy": 5.45648045539856,
      "epoch": 5.053066037735849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027442405635643574,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22948167026042937,
      "num_tokens": 115588747.0,
      "step": 55705
    },
    {
      "entropy": 5.477955102920532,
      "epoch": 5.053519593613934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002743917000487027,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22667172849178313,
      "num_tokens": 115599927.0,
      "step": 55710
    },
    {
      "entropy": 5.580211019515991,
      "epoch": 5.053973149492017,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00027435934375354934,
      "loss": 4.9485,
      "mean_token_accuracy": 0.22629450410604476,
      "num_tokens": 115610120.0,
      "step": 55715
    },
    {
      "entropy": 5.524288606643677,
      "epoch": 5.054426705370101,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00027432698747164496,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22317937463521959,
      "num_tokens": 115622039.0,
      "step": 55720
    },
    {
      "entropy": 5.473791027069092,
      "epoch": 5.054880261248186,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00027429463120365874,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2259483516216278,
      "num_tokens": 115633527.0,
      "step": 55725
    },
    {
      "entropy": 5.502677965164184,
      "epoch": 5.05533381712627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002742622749502596,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22534045726060867,
      "num_tokens": 115644363.0,
      "step": 55730
    },
    {
      "entropy": 5.588381290435791,
      "epoch": 5.055787373004354,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002742299187121168,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21487103402614594,
      "num_tokens": 115654313.0,
      "step": 55735
    },
    {
      "entropy": 5.515480709075928,
      "epoch": 5.056240928882438,
      "grad_norm": 1.25,
      "learning_rate": 0.00027419756248989955,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2348560392856598,
      "num_tokens": 115664724.0,
      "step": 55740
    },
    {
      "entropy": 5.431486797332764,
      "epoch": 5.056694484760523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002741652062842769,
      "loss": 4.829,
      "mean_token_accuracy": 0.22764795869588852,
      "num_tokens": 115674037.0,
      "step": 55745
    },
    {
      "entropy": 5.446970462799072,
      "epoch": 5.057148040638607,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000274132850095918,
      "loss": 4.8321,
      "mean_token_accuracy": 0.22858414947986602,
      "num_tokens": 115684431.0,
      "step": 55750
    },
    {
      "entropy": 5.483805370330811,
      "epoch": 5.057601596516691,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027410049392549197,
      "loss": 4.845,
      "mean_token_accuracy": 0.22666092216968536,
      "num_tokens": 115695293.0,
      "step": 55755
    },
    {
      "entropy": 5.4731680393219,
      "epoch": 5.058055152394775,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027406813777366794,
      "loss": 4.8372,
      "mean_token_accuracy": 0.23318780213594437,
      "num_tokens": 115705906.0,
      "step": 55760
    },
    {
      "entropy": 5.434492588043213,
      "epoch": 5.058508708272859,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027403578164111517,
      "loss": 4.8032,
      "mean_token_accuracy": 0.23598072826862335,
      "num_tokens": 115716157.0,
      "step": 55765
    },
    {
      "entropy": 5.563861846923828,
      "epoch": 5.058962264150943,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002740034255285025,
      "loss": 4.936,
      "mean_token_accuracy": 0.22489724755287172,
      "num_tokens": 115727691.0,
      "step": 55770
    },
    {
      "entropy": 5.494338130950927,
      "epoch": 5.059415820029027,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002739710694364994,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22231868803501129,
      "num_tokens": 115738984.0,
      "step": 55775
    },
    {
      "entropy": 5.447434663772583,
      "epoch": 5.059869375907112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027393871336577476,
      "loss": 4.8294,
      "mean_token_accuracy": 0.23095056414604187,
      "num_tokens": 115749040.0,
      "step": 55780
    },
    {
      "entropy": 5.491810321807861,
      "epoch": 5.060322931785196,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027390635731699774,
      "loss": 4.947,
      "mean_token_accuracy": 0.21895187199115754,
      "num_tokens": 115760125.0,
      "step": 55785
    },
    {
      "entropy": 5.47627067565918,
      "epoch": 5.06077648766328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002738740012908376,
      "loss": 4.874,
      "mean_token_accuracy": 0.23047596365213394,
      "num_tokens": 115770780.0,
      "step": 55790
    },
    {
      "entropy": 5.521477603912354,
      "epoch": 5.061230043541364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027384164528796345,
      "loss": 4.8462,
      "mean_token_accuracy": 0.23231585025787355,
      "num_tokens": 115782051.0,
      "step": 55795
    },
    {
      "entropy": 5.438287591934204,
      "epoch": 5.061683599419449,
      "grad_norm": 1.125,
      "learning_rate": 0.0002738092893090443,
      "loss": 4.7812,
      "mean_token_accuracy": 0.22909479588270187,
      "num_tokens": 115791692.0,
      "step": 55800
    },
    {
      "entropy": 5.487911939620972,
      "epoch": 5.062137155297533,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002737769333547494,
      "loss": 4.8143,
      "mean_token_accuracy": 0.24202586859464645,
      "num_tokens": 115802738.0,
      "step": 55805
    },
    {
      "entropy": 5.494489765167236,
      "epoch": 5.062590711175617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002737445774257478,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22635169625282286,
      "num_tokens": 115813073.0,
      "step": 55810
    },
    {
      "entropy": 5.522400999069214,
      "epoch": 5.063044267053701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002737122215227087,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2219857931137085,
      "num_tokens": 115823976.0,
      "step": 55815
    },
    {
      "entropy": 5.4305219650268555,
      "epoch": 5.063497822931785,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002736798656463011,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23166205286979674,
      "num_tokens": 115833332.0,
      "step": 55820
    },
    {
      "entropy": 5.4630420207977295,
      "epoch": 5.063951378809869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027364750979719433,
      "loss": 4.8399,
      "mean_token_accuracy": 0.22794530689716339,
      "num_tokens": 115844053.0,
      "step": 55825
    },
    {
      "entropy": 5.468730926513672,
      "epoch": 5.064404934687953,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002736151539760573,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2386629283428192,
      "num_tokens": 115854860.0,
      "step": 55830
    },
    {
      "entropy": 5.508822345733643,
      "epoch": 5.064858490566038,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027358279818355926,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23389069437980653,
      "num_tokens": 115865523.0,
      "step": 55835
    },
    {
      "entropy": 5.4615076065063475,
      "epoch": 5.065312046444122,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00027355044242036934,
      "loss": 4.8517,
      "mean_token_accuracy": 0.22743304520845414,
      "num_tokens": 115876033.0,
      "step": 55840
    },
    {
      "entropy": 5.48292989730835,
      "epoch": 5.065765602322206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002735180866871567,
      "loss": 4.8967,
      "mean_token_accuracy": 0.23154753893613816,
      "num_tokens": 115886575.0,
      "step": 55845
    },
    {
      "entropy": 5.4832391262054445,
      "epoch": 5.0662191582002905,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002734857309845903,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22083289921283722,
      "num_tokens": 115897628.0,
      "step": 55850
    },
    {
      "entropy": 5.46158766746521,
      "epoch": 5.066672714078375,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002734533753133394,
      "loss": 4.833,
      "mean_token_accuracy": 0.2306232124567032,
      "num_tokens": 115908472.0,
      "step": 55855
    },
    {
      "entropy": 5.4641814708709715,
      "epoch": 5.067126269956459,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027342101967407306,
      "loss": 4.7763,
      "mean_token_accuracy": 0.2351347789168358,
      "num_tokens": 115918798.0,
      "step": 55860
    },
    {
      "entropy": 5.44793586730957,
      "epoch": 5.067579825834543,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002733886640674605,
      "loss": 4.7956,
      "mean_token_accuracy": 0.23737668246030807,
      "num_tokens": 115928825.0,
      "step": 55865
    },
    {
      "entropy": 5.448257207870483,
      "epoch": 5.068033381712627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002733563084941707,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23168065696954726,
      "num_tokens": 115939315.0,
      "step": 55870
    },
    {
      "entropy": 5.442425966262817,
      "epoch": 5.068486937590711,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027332395295487296,
      "loss": 4.8774,
      "mean_token_accuracy": 0.23056268841028213,
      "num_tokens": 115949870.0,
      "step": 55875
    },
    {
      "entropy": 5.625319337844848,
      "epoch": 5.068940493468795,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027329159745023625,
      "loss": 5.0368,
      "mean_token_accuracy": 0.2131389483809471,
      "num_tokens": 115961108.0,
      "step": 55880
    },
    {
      "entropy": 5.499752235412598,
      "epoch": 5.069394049346879,
      "grad_norm": 1.015625,
      "learning_rate": 0.00027325924198092974,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22862233221530914,
      "num_tokens": 115972032.0,
      "step": 55885
    },
    {
      "entropy": 5.49223690032959,
      "epoch": 5.069847605224964,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002732268865476226,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2298046886920929,
      "num_tokens": 115982819.0,
      "step": 55890
    },
    {
      "entropy": 5.487604331970215,
      "epoch": 5.070301161103048,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002731945311509838,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22678643316030503,
      "num_tokens": 115992856.0,
      "step": 55895
    },
    {
      "entropy": 5.503676509857177,
      "epoch": 5.070754716981132,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002731621757916826,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22577383667230605,
      "num_tokens": 116004122.0,
      "step": 55900
    },
    {
      "entropy": 5.450528192520141,
      "epoch": 5.0712082728592165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000273129820470388,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23457521945238113,
      "num_tokens": 116014238.0,
      "step": 55905
    },
    {
      "entropy": 5.428070688247681,
      "epoch": 5.071661828737301,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027309746518776936,
      "loss": 4.7249,
      "mean_token_accuracy": 0.24074368476867675,
      "num_tokens": 116024310.0,
      "step": 55910
    },
    {
      "entropy": 5.5362883567810055,
      "epoch": 5.072115384615385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002730651099444955,
      "loss": 4.8692,
      "mean_token_accuracy": 0.23131955564022064,
      "num_tokens": 116034648.0,
      "step": 55915
    },
    {
      "entropy": 5.4782548427581785,
      "epoch": 5.072568940493468,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002730327547412357,
      "loss": 4.8729,
      "mean_token_accuracy": 0.23385726809501647,
      "num_tokens": 116045178.0,
      "step": 55920
    },
    {
      "entropy": 5.462580728530884,
      "epoch": 5.073022496371553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027300039957865903,
      "loss": 4.8841,
      "mean_token_accuracy": 0.23055093884468078,
      "num_tokens": 116056429.0,
      "step": 55925
    },
    {
      "entropy": 5.446298694610595,
      "epoch": 5.073476052249637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002729680444574346,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22902025282382965,
      "num_tokens": 116066151.0,
      "step": 55930
    },
    {
      "entropy": 5.498580598831177,
      "epoch": 5.073929608127721,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027293568937823156,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22942921370267869,
      "num_tokens": 116075330.0,
      "step": 55935
    },
    {
      "entropy": 5.509336233139038,
      "epoch": 5.0743831640058055,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002729033343417189,
      "loss": 4.9046,
      "mean_token_accuracy": 0.22563780397176741,
      "num_tokens": 116086108.0,
      "step": 55940
    },
    {
      "entropy": 5.528216171264648,
      "epoch": 5.07483671988389,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027287097934856593,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21621634662151337,
      "num_tokens": 116097046.0,
      "step": 55945
    },
    {
      "entropy": 5.465492486953735,
      "epoch": 5.075290275761974,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027283862439944157,
      "loss": 4.8348,
      "mean_token_accuracy": 0.2303678497672081,
      "num_tokens": 116108048.0,
      "step": 55950
    },
    {
      "entropy": 5.490974760055542,
      "epoch": 5.075743831640058,
      "grad_norm": 1.140625,
      "learning_rate": 0.000272806269495015,
      "loss": 4.8905,
      "mean_token_accuracy": 0.23065954595804214,
      "num_tokens": 116117417.0,
      "step": 55955
    },
    {
      "entropy": 5.611968517303467,
      "epoch": 5.0761973875181425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027277391463595543,
      "loss": 5.0101,
      "mean_token_accuracy": 0.21483954042196274,
      "num_tokens": 116128462.0,
      "step": 55960
    },
    {
      "entropy": 5.637499713897705,
      "epoch": 5.076650943396227,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002727415598229318,
      "loss": 5.0078,
      "mean_token_accuracy": 0.2200799584388733,
      "num_tokens": 116139908.0,
      "step": 55965
    },
    {
      "entropy": 5.529424524307251,
      "epoch": 5.07710449927431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002727092050566133,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2281218022108078,
      "num_tokens": 116149310.0,
      "step": 55970
    },
    {
      "entropy": 5.476081323623657,
      "epoch": 5.0775580551523944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000272676850337669,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2353903293609619,
      "num_tokens": 116160010.0,
      "step": 55975
    },
    {
      "entropy": 5.492408657073975,
      "epoch": 5.078011611030479,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002726444956667681,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2221398875117302,
      "num_tokens": 116170042.0,
      "step": 55980
    },
    {
      "entropy": 5.4947878360748295,
      "epoch": 5.078465166908563,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002726121410445796,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22786853164434434,
      "num_tokens": 116180842.0,
      "step": 55985
    },
    {
      "entropy": 5.562396764755249,
      "epoch": 5.078918722786647,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002725797864717726,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22330880910158157,
      "num_tokens": 116191496.0,
      "step": 55990
    },
    {
      "entropy": 5.501525592803955,
      "epoch": 5.0793722786647315,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002725474319490162,
      "loss": 4.8041,
      "mean_token_accuracy": 0.23502460420131682,
      "num_tokens": 116201451.0,
      "step": 55995
    },
    {
      "entropy": 5.475981760025024,
      "epoch": 5.079825834542816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002725150774769796,
      "loss": 4.7802,
      "mean_token_accuracy": 0.22916037142276763,
      "num_tokens": 116211327.0,
      "step": 56000
    },
    {
      "entropy": 5.474671459197998,
      "epoch": 5.0802793904209,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002724827230563319,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22639061510562897,
      "num_tokens": 116221676.0,
      "step": 56005
    },
    {
      "entropy": 5.453789806365966,
      "epoch": 5.080732946298984,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027245036868774205,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22505896240472795,
      "num_tokens": 116232778.0,
      "step": 56010
    },
    {
      "entropy": 5.433266353607178,
      "epoch": 5.081186502177069,
      "grad_norm": 1.0,
      "learning_rate": 0.0002724180143718792,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2363048419356346,
      "num_tokens": 116242198.0,
      "step": 56015
    },
    {
      "entropy": 5.419201946258545,
      "epoch": 5.081640058055152,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00027238566010941243,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23142455965280534,
      "num_tokens": 116252171.0,
      "step": 56020
    },
    {
      "entropy": 5.414696264266968,
      "epoch": 5.082093613933236,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000272353305901011,
      "loss": 4.7216,
      "mean_token_accuracy": 0.2391644224524498,
      "num_tokens": 116263014.0,
      "step": 56025
    },
    {
      "entropy": 5.4473076343536375,
      "epoch": 5.0825471698113205,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002723209517473438,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2277806043624878,
      "num_tokens": 116273126.0,
      "step": 56030
    },
    {
      "entropy": 5.504514265060425,
      "epoch": 5.083000725689405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027228859764908,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2278935819864273,
      "num_tokens": 116283174.0,
      "step": 56035
    },
    {
      "entropy": 5.476911163330078,
      "epoch": 5.083454281567489,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027225624360688864,
      "loss": 4.8809,
      "mean_token_accuracy": 0.23091951310634612,
      "num_tokens": 116293015.0,
      "step": 56040
    },
    {
      "entropy": 5.52295823097229,
      "epoch": 5.083907837445573,
      "grad_norm": 1.015625,
      "learning_rate": 0.000272223889621439,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2209248811006546,
      "num_tokens": 116304578.0,
      "step": 56045
    },
    {
      "entropy": 5.517996692657471,
      "epoch": 5.0843613933236576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002721915356933999,
      "loss": 4.8673,
      "mean_token_accuracy": 0.23098493814468385,
      "num_tokens": 116314666.0,
      "step": 56050
    },
    {
      "entropy": 5.487105369567871,
      "epoch": 5.084814949201742,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027215918182344064,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23604554682970047,
      "num_tokens": 116324972.0,
      "step": 56055
    },
    {
      "entropy": 5.4965660095214846,
      "epoch": 5.085268505079826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002721268280122302,
      "loss": 4.8494,
      "mean_token_accuracy": 0.22875131219625472,
      "num_tokens": 116335653.0,
      "step": 56060
    },
    {
      "entropy": 5.576427173614502,
      "epoch": 5.08572206095791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002720944742604377,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22449569702148436,
      "num_tokens": 116345522.0,
      "step": 56065
    },
    {
      "entropy": 5.451239681243896,
      "epoch": 5.086175616835995,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027206212056873213,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22646217942237853,
      "num_tokens": 116354889.0,
      "step": 56070
    },
    {
      "entropy": 5.391932201385498,
      "epoch": 5.086629172714078,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002720297669377828,
      "loss": 4.7833,
      "mean_token_accuracy": 0.2401527628302574,
      "num_tokens": 116364619.0,
      "step": 56075
    },
    {
      "entropy": 5.414196538925171,
      "epoch": 5.087082728592162,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027199741336825857,
      "loss": 4.7852,
      "mean_token_accuracy": 0.2369343563914299,
      "num_tokens": 116374346.0,
      "step": 56080
    },
    {
      "entropy": 5.501762628555298,
      "epoch": 5.0875362844702465,
      "grad_norm": 1.125,
      "learning_rate": 0.0002719650598608285,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2347957029938698,
      "num_tokens": 116385780.0,
      "step": 56085
    },
    {
      "entropy": 5.558188199996948,
      "epoch": 5.087989840348331,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002719327064161619,
      "loss": 4.8294,
      "mean_token_accuracy": 0.22802048474550246,
      "num_tokens": 116395232.0,
      "step": 56090
    },
    {
      "entropy": 5.461855792999268,
      "epoch": 5.088443396226415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002719003530349277,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23501128554344178,
      "num_tokens": 116404478.0,
      "step": 56095
    },
    {
      "entropy": 5.4526714324951175,
      "epoch": 5.088896952104499,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002718679997177949,
      "loss": 4.7909,
      "mean_token_accuracy": 0.23542621433734895,
      "num_tokens": 116414493.0,
      "step": 56100
    },
    {
      "entropy": 5.517755126953125,
      "epoch": 5.089350507982584,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027183564646543273,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2225555956363678,
      "num_tokens": 116424722.0,
      "step": 56105
    },
    {
      "entropy": 5.420002269744873,
      "epoch": 5.089804063860668,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027180329327851024,
      "loss": 4.7748,
      "mean_token_accuracy": 0.23009852021932603,
      "num_tokens": 116435639.0,
      "step": 56110
    },
    {
      "entropy": 5.422436237335205,
      "epoch": 5.090257619738752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027177094015769644,
      "loss": 4.7617,
      "mean_token_accuracy": 0.23146719336509705,
      "num_tokens": 116444908.0,
      "step": 56115
    },
    {
      "entropy": 5.497092294692993,
      "epoch": 5.090711175616836,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027173858710366033,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23626846671104432,
      "num_tokens": 116454965.0,
      "step": 56120
    },
    {
      "entropy": 5.459772443771362,
      "epoch": 5.09116473149492,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027170623411707114,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2238902196288109,
      "num_tokens": 116465861.0,
      "step": 56125
    },
    {
      "entropy": 5.491799211502075,
      "epoch": 5.091618287373004,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027167388119859787,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22186178267002105,
      "num_tokens": 116476367.0,
      "step": 56130
    },
    {
      "entropy": 5.4571967124938965,
      "epoch": 5.092071843251088,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027164152834890953,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22963497489690782,
      "num_tokens": 116486492.0,
      "step": 56135
    },
    {
      "entropy": 5.452378463745117,
      "epoch": 5.092525399129173,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027160917556867534,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2300916463136673,
      "num_tokens": 116496280.0,
      "step": 56140
    },
    {
      "entropy": 5.556349563598633,
      "epoch": 5.092978955007257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002715768228585642,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21681967824697496,
      "num_tokens": 116505942.0,
      "step": 56145
    },
    {
      "entropy": 5.525787782669068,
      "epoch": 5.093432510885341,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002715444702192453,
      "loss": 4.8216,
      "mean_token_accuracy": 0.22624819725751877,
      "num_tokens": 116515749.0,
      "step": 56150
    },
    {
      "entropy": 5.46392970085144,
      "epoch": 5.093886066763425,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002715121176513875,
      "loss": 4.8401,
      "mean_token_accuracy": 0.23149076700210572,
      "num_tokens": 116525569.0,
      "step": 56155
    },
    {
      "entropy": 5.574133777618409,
      "epoch": 5.09433962264151,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002714797651556601,
      "loss": 4.9653,
      "mean_token_accuracy": 0.22791660875082015,
      "num_tokens": 116535837.0,
      "step": 56160
    },
    {
      "entropy": 5.4833110809326175,
      "epoch": 5.094793178519594,
      "grad_norm": 1.125,
      "learning_rate": 0.00027144741273273207,
      "loss": 4.8496,
      "mean_token_accuracy": 0.223638354241848,
      "num_tokens": 116546299.0,
      "step": 56165
    },
    {
      "entropy": 5.5199951171875,
      "epoch": 5.095246734397678,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027141506038327233,
      "loss": 4.848,
      "mean_token_accuracy": 0.23336009532213212,
      "num_tokens": 116556725.0,
      "step": 56170
    },
    {
      "entropy": 5.455988836288452,
      "epoch": 5.0957002902757615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002713827081079502,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22782676219940184,
      "num_tokens": 116566853.0,
      "step": 56175
    },
    {
      "entropy": 5.465390634536743,
      "epoch": 5.096153846153846,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002713503559074345,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2265086442232132,
      "num_tokens": 116578347.0,
      "step": 56180
    },
    {
      "entropy": 5.510306072235108,
      "epoch": 5.09660740203193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027131800378239437,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21865545809268952,
      "num_tokens": 116588836.0,
      "step": 56185
    },
    {
      "entropy": 5.482001781463623,
      "epoch": 5.097060957910014,
      "grad_norm": 1.125,
      "learning_rate": 0.0002712856517334989,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23717603087425232,
      "num_tokens": 116599463.0,
      "step": 56190
    },
    {
      "entropy": 5.519666337966919,
      "epoch": 5.097514513788099,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027125329976141713,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2173708513379097,
      "num_tokens": 116610329.0,
      "step": 56195
    },
    {
      "entropy": 5.517553043365479,
      "epoch": 5.097968069666183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000271220947866818,
      "loss": 4.8574,
      "mean_token_accuracy": 0.23580856174230574,
      "num_tokens": 116619954.0,
      "step": 56200
    },
    {
      "entropy": 5.5340869426727295,
      "epoch": 5.098421625544267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002711885960503706,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23489078134298325,
      "num_tokens": 116629347.0,
      "step": 56205
    },
    {
      "entropy": 5.499699449539184,
      "epoch": 5.098875181422351,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002711562443127441,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23236748725175857,
      "num_tokens": 116639674.0,
      "step": 56210
    },
    {
      "entropy": 5.471911668777466,
      "epoch": 5.099328737300436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002711238926546074,
      "loss": 4.8234,
      "mean_token_accuracy": 0.22269484549760818,
      "num_tokens": 116650889.0,
      "step": 56215
    },
    {
      "entropy": 5.38556547164917,
      "epoch": 5.09978229317852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027109154107662957,
      "loss": 4.834,
      "mean_token_accuracy": 0.22296947091817856,
      "num_tokens": 116661213.0,
      "step": 56220
    },
    {
      "entropy": 5.457258224487305,
      "epoch": 5.100235849056604,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002710591895794797,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22999518364667892,
      "num_tokens": 116672064.0,
      "step": 56225
    },
    {
      "entropy": 5.548843574523926,
      "epoch": 5.100689404934688,
      "grad_norm": 1.125,
      "learning_rate": 0.0002710268381638268,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2194358304142952,
      "num_tokens": 116681616.0,
      "step": 56230
    },
    {
      "entropy": 5.47413387298584,
      "epoch": 5.101142960812772,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002709944868303399,
      "loss": 4.7999,
      "mean_token_accuracy": 0.23446125239133836,
      "num_tokens": 116691427.0,
      "step": 56235
    },
    {
      "entropy": 5.460171747207641,
      "epoch": 5.101596516690856,
      "grad_norm": 1.3125,
      "learning_rate": 0.000270962135579688,
      "loss": 4.8453,
      "mean_token_accuracy": 0.22925375699996947,
      "num_tokens": 116701326.0,
      "step": 56240
    },
    {
      "entropy": 5.4671379089355465,
      "epoch": 5.10205007256894,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002709297844125402,
      "loss": 4.886,
      "mean_token_accuracy": 0.2260168746113777,
      "num_tokens": 116712353.0,
      "step": 56245
    },
    {
      "entropy": 5.507921266555786,
      "epoch": 5.102503628447025,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002708974333295655,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23121041506528855,
      "num_tokens": 116721906.0,
      "step": 56250
    },
    {
      "entropy": 5.510182571411133,
      "epoch": 5.102957184325109,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027086508233143287,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2263959303498268,
      "num_tokens": 116732684.0,
      "step": 56255
    },
    {
      "entropy": 5.473689413070678,
      "epoch": 5.103410740203193,
      "grad_norm": 1.0,
      "learning_rate": 0.0002708327314188114,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2293819457292557,
      "num_tokens": 116743071.0,
      "step": 56260
    },
    {
      "entropy": 5.4902770042419435,
      "epoch": 5.1038642960812775,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002708003805923702,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2249564126133919,
      "num_tokens": 116752756.0,
      "step": 56265
    },
    {
      "entropy": 5.477485704421997,
      "epoch": 5.104317851959362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027076802985277813,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2287900224328041,
      "num_tokens": 116762878.0,
      "step": 56270
    },
    {
      "entropy": 5.539960336685181,
      "epoch": 5.104771407837446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002707356792007042,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22857471108436583,
      "num_tokens": 116772953.0,
      "step": 56275
    },
    {
      "entropy": 5.585286092758179,
      "epoch": 5.105224963715529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027070332863681764,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22756002545356752,
      "num_tokens": 116783530.0,
      "step": 56280
    },
    {
      "entropy": 5.477519798278808,
      "epoch": 5.105678519593614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027067097816178734,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22884192019701005,
      "num_tokens": 116792740.0,
      "step": 56285
    },
    {
      "entropy": 5.448164558410644,
      "epoch": 5.106132075471698,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002706386277762822,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2331846386194229,
      "num_tokens": 116802846.0,
      "step": 56290
    },
    {
      "entropy": 5.489220476150512,
      "epoch": 5.106585631349782,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002706062774809714,
      "loss": 4.8978,
      "mean_token_accuracy": 0.2220088005065918,
      "num_tokens": 116812787.0,
      "step": 56295
    },
    {
      "entropy": 5.547648143768311,
      "epoch": 5.107039187227866,
      "grad_norm": 1.015625,
      "learning_rate": 0.000270573927276524,
      "loss": 4.9101,
      "mean_token_accuracy": 0.224932861328125,
      "num_tokens": 116823840.0,
      "step": 56300
    },
    {
      "entropy": 5.506267452239991,
      "epoch": 5.107492743105951,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002705415771636088,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23016982078552245,
      "num_tokens": 116833971.0,
      "step": 56305
    },
    {
      "entropy": 5.441998386383057,
      "epoch": 5.107946298984035,
      "grad_norm": 1.125,
      "learning_rate": 0.00027050922714289493,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23019986003637313,
      "num_tokens": 116844292.0,
      "step": 56310
    },
    {
      "entropy": 5.410767889022827,
      "epoch": 5.108399854862119,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027047687721505143,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22261336743831633,
      "num_tokens": 116854399.0,
      "step": 56315
    },
    {
      "entropy": 5.490280437469482,
      "epoch": 5.1088534107402035,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027044452738074726,
      "loss": 4.8361,
      "mean_token_accuracy": 0.22880103141069413,
      "num_tokens": 116865418.0,
      "step": 56320
    },
    {
      "entropy": 5.428809642791748,
      "epoch": 5.109306966618288,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027041217764065136,
      "loss": 4.8112,
      "mean_token_accuracy": 0.23614969700574875,
      "num_tokens": 116875225.0,
      "step": 56325
    },
    {
      "entropy": 5.536821365356445,
      "epoch": 5.109760522496371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027037982799543286,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22389578819274902,
      "num_tokens": 116885062.0,
      "step": 56330
    },
    {
      "entropy": 5.525861215591431,
      "epoch": 5.110214078374455,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002703474784457607,
      "loss": 4.9348,
      "mean_token_accuracy": 0.22242972552776336,
      "num_tokens": 116894705.0,
      "step": 56335
    },
    {
      "entropy": 5.483656597137451,
      "epoch": 5.11066763425254,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027031512899230374,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22420355230569838,
      "num_tokens": 116904441.0,
      "step": 56340
    },
    {
      "entropy": 5.4660196781158445,
      "epoch": 5.111121190130624,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002702827796357313,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23090738654136658,
      "num_tokens": 116915008.0,
      "step": 56345
    },
    {
      "entropy": 5.543429756164551,
      "epoch": 5.111574746008708,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002702504303767121,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22174298614263535,
      "num_tokens": 116926035.0,
      "step": 56350
    },
    {
      "entropy": 5.545201826095581,
      "epoch": 5.1120283018867925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002702180812159152,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2340882584452629,
      "num_tokens": 116936263.0,
      "step": 56355
    },
    {
      "entropy": 5.475631618499756,
      "epoch": 5.112481857764877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027018573215400955,
      "loss": 4.7307,
      "mean_token_accuracy": 0.23648300021886826,
      "num_tokens": 116947080.0,
      "step": 56360
    },
    {
      "entropy": 5.49696979522705,
      "epoch": 5.112935413642961,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027015338319166424,
      "loss": 4.7979,
      "mean_token_accuracy": 0.22975894808769226,
      "num_tokens": 116958224.0,
      "step": 56365
    },
    {
      "entropy": 5.452006816864014,
      "epoch": 5.113388969521045,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002701210343295482,
      "loss": 4.907,
      "mean_token_accuracy": 0.22704071402549744,
      "num_tokens": 116968385.0,
      "step": 56370
    },
    {
      "entropy": 5.506493043899536,
      "epoch": 5.1138425253991295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027008868556833044,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22221098989248275,
      "num_tokens": 116979309.0,
      "step": 56375
    },
    {
      "entropy": 5.5518053531646725,
      "epoch": 5.114296081277214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027005633690867985,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2279647946357727,
      "num_tokens": 116990739.0,
      "step": 56380
    },
    {
      "entropy": 5.523322820663452,
      "epoch": 5.114749637155297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002700239883512655,
      "loss": 4.8807,
      "mean_token_accuracy": 0.22579692155122758,
      "num_tokens": 117000820.0,
      "step": 56385
    },
    {
      "entropy": 5.4227947235107425,
      "epoch": 5.1152031930333814,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002699916398967563,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23229169696569443,
      "num_tokens": 117011319.0,
      "step": 56390
    },
    {
      "entropy": 5.498776960372925,
      "epoch": 5.115656748911466,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002699592915458214,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23014702647924423,
      "num_tokens": 117021480.0,
      "step": 56395
    },
    {
      "entropy": 5.563027667999267,
      "epoch": 5.11611030478955,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00026992694329912954,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22114156782627106,
      "num_tokens": 117033237.0,
      "step": 56400
    },
    {
      "entropy": 5.496589612960816,
      "epoch": 5.116563860667634,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026989459515734974,
      "loss": 4.825,
      "mean_token_accuracy": 0.23188442140817642,
      "num_tokens": 117044237.0,
      "step": 56405
    },
    {
      "entropy": 5.494913864135742,
      "epoch": 5.1170174165457185,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026986224712115107,
      "loss": 4.8435,
      "mean_token_accuracy": 0.2290666714310646,
      "num_tokens": 117054783.0,
      "step": 56410
    },
    {
      "entropy": 5.4722129821777346,
      "epoch": 5.117470972423803,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002698298991912024,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22645662873983383,
      "num_tokens": 117064865.0,
      "step": 56415
    },
    {
      "entropy": 5.506269216537476,
      "epoch": 5.117924528301887,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002697975513681728,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22847946584224701,
      "num_tokens": 117074968.0,
      "step": 56420
    },
    {
      "entropy": 5.489071893692016,
      "epoch": 5.118378084179971,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002697652036527311,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22893763482570648,
      "num_tokens": 117086574.0,
      "step": 56425
    },
    {
      "entropy": 5.4256591796875,
      "epoch": 5.118831640058056,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026973285604554647,
      "loss": 4.6971,
      "mean_token_accuracy": 0.24441557824611665,
      "num_tokens": 117097212.0,
      "step": 56430
    },
    {
      "entropy": 5.488743591308594,
      "epoch": 5.119285195936139,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026970050854728756,
      "loss": 4.7639,
      "mean_token_accuracy": 0.23828944116830825,
      "num_tokens": 117107544.0,
      "step": 56435
    },
    {
      "entropy": 5.518773603439331,
      "epoch": 5.119738751814223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026966816115862354,
      "loss": 4.9381,
      "mean_token_accuracy": 0.21747301518917084,
      "num_tokens": 117118300.0,
      "step": 56440
    },
    {
      "entropy": 5.50610842704773,
      "epoch": 5.1201923076923075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002696358138802233,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23639891743659974,
      "num_tokens": 117128702.0,
      "step": 56445
    },
    {
      "entropy": 5.4035194396972654,
      "epoch": 5.120645863570392,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026960346671275583,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2354031592607498,
      "num_tokens": 117139485.0,
      "step": 56450
    },
    {
      "entropy": 5.560116863250732,
      "epoch": 5.121099419448476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026957111965688994,
      "loss": 4.9377,
      "mean_token_accuracy": 0.2256823107600212,
      "num_tokens": 117149895.0,
      "step": 56455
    },
    {
      "entropy": 5.50617241859436,
      "epoch": 5.12155297532656,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026953877271329476,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22639549374580384,
      "num_tokens": 117160901.0,
      "step": 56460
    },
    {
      "entropy": 5.433061981201172,
      "epoch": 5.1220065312046446,
      "grad_norm": 1.125,
      "learning_rate": 0.00026950642588263914,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2388167217373848,
      "num_tokens": 117170820.0,
      "step": 56465
    },
    {
      "entropy": 5.448678398132325,
      "epoch": 5.122460087082729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026947407916559196,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23067855834960938,
      "num_tokens": 117181581.0,
      "step": 56470
    },
    {
      "entropy": 5.4480798721313475,
      "epoch": 5.122913642960813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026944173256282234,
      "loss": 4.7911,
      "mean_token_accuracy": 0.23371865451335908,
      "num_tokens": 117190817.0,
      "step": 56475
    },
    {
      "entropy": 5.524083852767944,
      "epoch": 5.123367198838897,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026940938607499903,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2309391140937805,
      "num_tokens": 117201459.0,
      "step": 56480
    },
    {
      "entropy": 5.4316507339477536,
      "epoch": 5.123820754716981,
      "grad_norm": 1.125,
      "learning_rate": 0.0002693770397027911,
      "loss": 4.7916,
      "mean_token_accuracy": 0.23016009330749512,
      "num_tokens": 117211941.0,
      "step": 56485
    },
    {
      "entropy": 5.478974914550781,
      "epoch": 5.124274310595065,
      "grad_norm": 1.0,
      "learning_rate": 0.00026934469344686734,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2225956693291664,
      "num_tokens": 117222010.0,
      "step": 56490
    },
    {
      "entropy": 5.494577407836914,
      "epoch": 5.124727866473149,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002693123473078968,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22224222272634506,
      "num_tokens": 117231453.0,
      "step": 56495
    },
    {
      "entropy": 5.512209701538086,
      "epoch": 5.1251814223512335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026928000128654845,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22879091203212737,
      "num_tokens": 117241348.0,
      "step": 56500
    },
    {
      "entropy": 5.485125637054443,
      "epoch": 5.125634978229318,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000269247655383491,
      "loss": 4.8096,
      "mean_token_accuracy": 0.22912387251853944,
      "num_tokens": 117251472.0,
      "step": 56505
    },
    {
      "entropy": 5.461377334594727,
      "epoch": 5.126088534107402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026921530959939354,
      "loss": 4.8226,
      "mean_token_accuracy": 0.23111433237791063,
      "num_tokens": 117262007.0,
      "step": 56510
    },
    {
      "entropy": 5.433710241317749,
      "epoch": 5.126542089985486,
      "grad_norm": 1.125,
      "learning_rate": 0.000269182963934925,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22777406871318817,
      "num_tokens": 117272573.0,
      "step": 56515
    },
    {
      "entropy": 5.498913097381592,
      "epoch": 5.126995645863571,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002691506183907542,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22558122277259826,
      "num_tokens": 117283832.0,
      "step": 56520
    },
    {
      "entropy": 5.530829048156738,
      "epoch": 5.127449201741655,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026911827296755014,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2279312327504158,
      "num_tokens": 117295287.0,
      "step": 56525
    },
    {
      "entropy": 5.478460693359375,
      "epoch": 5.127902757619739,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002690859276659817,
      "loss": 4.8412,
      "mean_token_accuracy": 0.23257199227809905,
      "num_tokens": 117305585.0,
      "step": 56530
    },
    {
      "entropy": 5.542371654510498,
      "epoch": 5.1283563134978225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002690535824867178,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22611751407384872,
      "num_tokens": 117316149.0,
      "step": 56535
    },
    {
      "entropy": 5.495818853378296,
      "epoch": 5.128809869375907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002690212374304273,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23492420017719268,
      "num_tokens": 117325777.0,
      "step": 56540
    },
    {
      "entropy": 5.5237147331237795,
      "epoch": 5.129263425253991,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002689888924977791,
      "loss": 4.9505,
      "mean_token_accuracy": 0.22495925426483154,
      "num_tokens": 117336303.0,
      "step": 56545
    },
    {
      "entropy": 5.457941961288452,
      "epoch": 5.129716981132075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002689565476894422,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2342174008488655,
      "num_tokens": 117347169.0,
      "step": 56550
    },
    {
      "entropy": 5.457670068740844,
      "epoch": 5.13017053701016,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026892420300608533,
      "loss": 4.8197,
      "mean_token_accuracy": 0.24016400724649428,
      "num_tokens": 117357461.0,
      "step": 56555
    },
    {
      "entropy": 5.432500505447388,
      "epoch": 5.130624092888244,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026889185844837754,
      "loss": 4.888,
      "mean_token_accuracy": 0.22846860140562059,
      "num_tokens": 117367915.0,
      "step": 56560
    },
    {
      "entropy": 5.4925731182098385,
      "epoch": 5.131077648766328,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002688595140169877,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22291503548622132,
      "num_tokens": 117378352.0,
      "step": 56565
    },
    {
      "entropy": 5.560561323165894,
      "epoch": 5.131531204644412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002688271697125847,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22265047132968901,
      "num_tokens": 117389074.0,
      "step": 56570
    },
    {
      "entropy": 5.560022449493408,
      "epoch": 5.131984760522497,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002687948255358373,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22844196557998658,
      "num_tokens": 117399643.0,
      "step": 56575
    },
    {
      "entropy": 5.473238945007324,
      "epoch": 5.132438316400581,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026876248148741464,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23247677236795425,
      "num_tokens": 117410015.0,
      "step": 56580
    },
    {
      "entropy": 5.424787616729736,
      "epoch": 5.132891872278664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026873013756798533,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23305527120828629,
      "num_tokens": 117419517.0,
      "step": 56585
    },
    {
      "entropy": 5.3921332359313965,
      "epoch": 5.1333454281567485,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026869779377821837,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23447399884462355,
      "num_tokens": 117430450.0,
      "step": 56590
    },
    {
      "entropy": 5.472375917434692,
      "epoch": 5.133798984034833,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002686654501187827,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22757003903388978,
      "num_tokens": 117440403.0,
      "step": 56595
    },
    {
      "entropy": 5.462691354751587,
      "epoch": 5.134252539912917,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002686331065903471,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22384418547153473,
      "num_tokens": 117451014.0,
      "step": 56600
    },
    {
      "entropy": 5.534164381027222,
      "epoch": 5.134706095791001,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002686007631935805,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22494083195924758,
      "num_tokens": 117461238.0,
      "step": 56605
    },
    {
      "entropy": 5.540903377532959,
      "epoch": 5.135159651669086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002685684199291518,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22772126793861389,
      "num_tokens": 117472298.0,
      "step": 56610
    },
    {
      "entropy": 5.529396820068359,
      "epoch": 5.13561320754717,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026853607679772973,
      "loss": 4.9258,
      "mean_token_accuracy": 0.22898186296224593,
      "num_tokens": 117483010.0,
      "step": 56615
    },
    {
      "entropy": 5.370754051208496,
      "epoch": 5.136066763425254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026850373379998325,
      "loss": 4.7811,
      "mean_token_accuracy": 0.23527880012989044,
      "num_tokens": 117494325.0,
      "step": 56620
    },
    {
      "entropy": 5.516162967681884,
      "epoch": 5.136520319303338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026847139093658123,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2232827439904213,
      "num_tokens": 117505426.0,
      "step": 56625
    },
    {
      "entropy": 5.530826807022095,
      "epoch": 5.136973875181423,
      "grad_norm": 1.125,
      "learning_rate": 0.00026843904820819247,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22508124858140946,
      "num_tokens": 117515920.0,
      "step": 56630
    },
    {
      "entropy": 5.443610715866089,
      "epoch": 5.137427431059507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026840670561548597,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2264670178294182,
      "num_tokens": 117525454.0,
      "step": 56635
    },
    {
      "entropy": 5.421769714355468,
      "epoch": 5.13788098693759,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002683743631591303,
      "loss": 4.8035,
      "mean_token_accuracy": 0.23710936605930327,
      "num_tokens": 117536234.0,
      "step": 56640
    },
    {
      "entropy": 5.4915111541748045,
      "epoch": 5.138334542815675,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026834202083979465,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22199353128671645,
      "num_tokens": 117546639.0,
      "step": 56645
    },
    {
      "entropy": 5.48482232093811,
      "epoch": 5.138788098693759,
      "grad_norm": 1.0,
      "learning_rate": 0.00026830967865814763,
      "loss": 4.843,
      "mean_token_accuracy": 0.22510099411010742,
      "num_tokens": 117557104.0,
      "step": 56650
    },
    {
      "entropy": 5.563954830169678,
      "epoch": 5.139241654571843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026827733661485825,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2299998551607132,
      "num_tokens": 117568909.0,
      "step": 56655
    },
    {
      "entropy": 5.498415517807007,
      "epoch": 5.139695210449927,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026824499471059514,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22433033287525178,
      "num_tokens": 117579987.0,
      "step": 56660
    },
    {
      "entropy": 5.41833610534668,
      "epoch": 5.140148766328012,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026821265294602736,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22560183852910995,
      "num_tokens": 117591152.0,
      "step": 56665
    },
    {
      "entropy": 5.476552534103393,
      "epoch": 5.140602322206096,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002681803113218236,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23544468283653258,
      "num_tokens": 117601219.0,
      "step": 56670
    },
    {
      "entropy": 5.57867751121521,
      "epoch": 5.14105587808418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002681479698386527,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22909507751464844,
      "num_tokens": 117611670.0,
      "step": 56675
    },
    {
      "entropy": 5.490227317810058,
      "epoch": 5.1415094339622645,
      "grad_norm": 1.125,
      "learning_rate": 0.0002681156284971836,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2304386168718338,
      "num_tokens": 117622016.0,
      "step": 56680
    },
    {
      "entropy": 5.480059337615967,
      "epoch": 5.141962989840349,
      "grad_norm": 1.078125,
      "learning_rate": 0.000268083287298085,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2277246743440628,
      "num_tokens": 117631891.0,
      "step": 56685
    },
    {
      "entropy": 5.417768669128418,
      "epoch": 5.142416545718432,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002680509462420258,
      "loss": 4.7805,
      "mean_token_accuracy": 0.2383338764309883,
      "num_tokens": 117642333.0,
      "step": 56690
    },
    {
      "entropy": 5.415923595428467,
      "epoch": 5.142870101596516,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00026801860532967486,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23592499047517776,
      "num_tokens": 117652761.0,
      "step": 56695
    },
    {
      "entropy": 5.465116310119629,
      "epoch": 5.143323657474601,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002679862645617009,
      "loss": 4.81,
      "mean_token_accuracy": 0.23945456445217134,
      "num_tokens": 117661660.0,
      "step": 56700
    },
    {
      "entropy": 5.476190042495728,
      "epoch": 5.143777213352685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002679539239387728,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2226606860756874,
      "num_tokens": 117672083.0,
      "step": 56705
    },
    {
      "entropy": 5.437607860565185,
      "epoch": 5.144230769230769,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026792158346155924,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2367209479212761,
      "num_tokens": 117682923.0,
      "step": 56710
    },
    {
      "entropy": 5.423923301696777,
      "epoch": 5.144684325108853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026788924313072925,
      "loss": 4.8089,
      "mean_token_accuracy": 0.22793384045362472,
      "num_tokens": 117694409.0,
      "step": 56715
    },
    {
      "entropy": 5.508266353607178,
      "epoch": 5.145137880986938,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026785690294695155,
      "loss": 4.8656,
      "mean_token_accuracy": 0.23127015829086303,
      "num_tokens": 117704185.0,
      "step": 56720
    },
    {
      "entropy": 5.562346410751343,
      "epoch": 5.145591436865022,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026782456291089477,
      "loss": 4.9,
      "mean_token_accuracy": 0.2196995347738266,
      "num_tokens": 117715067.0,
      "step": 56725
    },
    {
      "entropy": 5.44383807182312,
      "epoch": 5.146044992743106,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026779222302322796,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22968607693910598,
      "num_tokens": 117724258.0,
      "step": 56730
    },
    {
      "entropy": 5.540481662750244,
      "epoch": 5.1464985486211905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002677598832846198,
      "loss": 4.9262,
      "mean_token_accuracy": 0.22435338497161866,
      "num_tokens": 117733334.0,
      "step": 56735
    },
    {
      "entropy": 5.512028312683105,
      "epoch": 5.146952104499274,
      "grad_norm": 1.0,
      "learning_rate": 0.00026772754369573914,
      "loss": 4.8783,
      "mean_token_accuracy": 0.23322980999946594,
      "num_tokens": 117744482.0,
      "step": 56740
    },
    {
      "entropy": 5.551439428329468,
      "epoch": 5.147405660377358,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026769520425725464,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22447903603315353,
      "num_tokens": 117755313.0,
      "step": 56745
    },
    {
      "entropy": 5.438523244857788,
      "epoch": 5.147859216255442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026766286496983523,
      "loss": 4.868,
      "mean_token_accuracy": 0.22663439065217972,
      "num_tokens": 117765892.0,
      "step": 56750
    },
    {
      "entropy": 5.456478452682495,
      "epoch": 5.148312772133527,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00026763052583414964,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22989501655101777,
      "num_tokens": 117776132.0,
      "step": 56755
    },
    {
      "entropy": 5.517182397842407,
      "epoch": 5.148766328011611,
      "grad_norm": 1.125,
      "learning_rate": 0.0002675981868508666,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2303432270884514,
      "num_tokens": 117787194.0,
      "step": 56760
    },
    {
      "entropy": 5.472425413131714,
      "epoch": 5.149219883889695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026756584802065504,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23028497099876405,
      "num_tokens": 117797991.0,
      "step": 56765
    },
    {
      "entropy": 5.52309021949768,
      "epoch": 5.1496734397677795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002675335093441836,
      "loss": 4.9472,
      "mean_token_accuracy": 0.22225910574197769,
      "num_tokens": 117808114.0,
      "step": 56770
    },
    {
      "entropy": 5.466439056396484,
      "epoch": 5.150126995645864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002675011708221211,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23220661282539368,
      "num_tokens": 117818413.0,
      "step": 56775
    },
    {
      "entropy": 5.4546839714050295,
      "epoch": 5.150580551523948,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026746883245513627,
      "loss": 4.794,
      "mean_token_accuracy": 0.2263038620352745,
      "num_tokens": 117828926.0,
      "step": 56780
    },
    {
      "entropy": 5.496087265014649,
      "epoch": 5.151034107402032,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002674364942438979,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2299483522772789,
      "num_tokens": 117841051.0,
      "step": 56785
    },
    {
      "entropy": 5.4066322326660154,
      "epoch": 5.1514876632801165,
      "grad_norm": 1.125,
      "learning_rate": 0.00026740415618907475,
      "loss": 4.8317,
      "mean_token_accuracy": 0.23232903182506562,
      "num_tokens": 117852010.0,
      "step": 56790
    },
    {
      "entropy": 5.511244201660157,
      "epoch": 5.1519412191582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002673718182913356,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22661216109991072,
      "num_tokens": 117860873.0,
      "step": 56795
    },
    {
      "entropy": 5.504572534561158,
      "epoch": 5.152394775036284,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002673394805513492,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2262418180704117,
      "num_tokens": 117871417.0,
      "step": 56800
    },
    {
      "entropy": 5.474633455276489,
      "epoch": 5.1528483309143684,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026730714296978433,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2332349583506584,
      "num_tokens": 117881115.0,
      "step": 56805
    },
    {
      "entropy": 5.508223342895508,
      "epoch": 5.153301886792453,
      "grad_norm": 1.125,
      "learning_rate": 0.0002672748055473096,
      "loss": 4.8328,
      "mean_token_accuracy": 0.23007166832685472,
      "num_tokens": 117891341.0,
      "step": 56810
    },
    {
      "entropy": 5.4639747619628904,
      "epoch": 5.153755442670537,
      "grad_norm": 1.28125,
      "learning_rate": 0.000267242468284594,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22580323368310928,
      "num_tokens": 117902304.0,
      "step": 56815
    },
    {
      "entropy": 5.408119010925293,
      "epoch": 5.154208998548621,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026721013118230605,
      "loss": 4.8168,
      "mean_token_accuracy": 0.23008167296648024,
      "num_tokens": 117913367.0,
      "step": 56820
    },
    {
      "entropy": 5.437274503707886,
      "epoch": 5.1546625544267055,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026717779424111456,
      "loss": 4.786,
      "mean_token_accuracy": 0.22990192770957946,
      "num_tokens": 117922750.0,
      "step": 56825
    },
    {
      "entropy": 5.5037088871002195,
      "epoch": 5.15511611030479,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026714545746168826,
      "loss": 4.835,
      "mean_token_accuracy": 0.22655624896287918,
      "num_tokens": 117933950.0,
      "step": 56830
    },
    {
      "entropy": 5.482941007614135,
      "epoch": 5.155569666182874,
      "grad_norm": 1.296875,
      "learning_rate": 0.000267113120844696,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2236953094601631,
      "num_tokens": 117944985.0,
      "step": 56835
    },
    {
      "entropy": 5.407291316986084,
      "epoch": 5.156023222060958,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002670807843908064,
      "loss": 4.7399,
      "mean_token_accuracy": 0.23936171680688859,
      "num_tokens": 117954560.0,
      "step": 56840
    },
    {
      "entropy": 5.438097238540649,
      "epoch": 5.156476777939042,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026704844810068805,
      "loss": 4.8193,
      "mean_token_accuracy": 0.23233485370874404,
      "num_tokens": 117965690.0,
      "step": 56845
    },
    {
      "entropy": 5.468141508102417,
      "epoch": 5.156930333817126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002670161119750099,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22612619400024414,
      "num_tokens": 117975856.0,
      "step": 56850
    },
    {
      "entropy": 5.4913379669189455,
      "epoch": 5.15738388969521,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002669837760144407,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2298653766512871,
      "num_tokens": 117985673.0,
      "step": 56855
    },
    {
      "entropy": 5.467463779449463,
      "epoch": 5.1578374455732945,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026695144021964895,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23147682696580887,
      "num_tokens": 117995866.0,
      "step": 56860
    },
    {
      "entropy": 5.465226125717163,
      "epoch": 5.158291001451379,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026691910459130346,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22390212416648864,
      "num_tokens": 118006756.0,
      "step": 56865
    },
    {
      "entropy": 5.503367853164673,
      "epoch": 5.158744557329463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026688676913007303,
      "loss": 4.94,
      "mean_token_accuracy": 0.22750447541475297,
      "num_tokens": 118016506.0,
      "step": 56870
    },
    {
      "entropy": 5.536367130279541,
      "epoch": 5.159198113207547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026685443383662615,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22742126286029815,
      "num_tokens": 118026681.0,
      "step": 56875
    },
    {
      "entropy": 5.473202705383301,
      "epoch": 5.1596516690856316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026682209871163175,
      "loss": 4.8518,
      "mean_token_accuracy": 0.22868201583623887,
      "num_tokens": 118036376.0,
      "step": 56880
    },
    {
      "entropy": 5.53194580078125,
      "epoch": 5.160105224963716,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002667897637557584,
      "loss": 4.8999,
      "mean_token_accuracy": 0.228643736243248,
      "num_tokens": 118047029.0,
      "step": 56885
    },
    {
      "entropy": 5.480251741409302,
      "epoch": 5.1605587808418,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002667574289696748,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22961509376764297,
      "num_tokens": 118056887.0,
      "step": 56890
    },
    {
      "entropy": 5.49562783241272,
      "epoch": 5.1610123367198835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026672509435404974,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22439484298229218,
      "num_tokens": 118067036.0,
      "step": 56895
    },
    {
      "entropy": 5.482414722442627,
      "epoch": 5.161465892597968,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026669275990955175,
      "loss": 4.8518,
      "mean_token_accuracy": 0.23903809487819672,
      "num_tokens": 118077439.0,
      "step": 56900
    },
    {
      "entropy": 5.5078040599823,
      "epoch": 5.161919448476052,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026666042563684966,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22054371386766433,
      "num_tokens": 118089009.0,
      "step": 56905
    },
    {
      "entropy": 5.500152683258056,
      "epoch": 5.162373004354136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002666280915366121,
      "loss": 4.884,
      "mean_token_accuracy": 0.22648645043373108,
      "num_tokens": 118098820.0,
      "step": 56910
    },
    {
      "entropy": 5.476612997055054,
      "epoch": 5.1628265602322205,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026659575760950774,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2325211673974991,
      "num_tokens": 118108896.0,
      "step": 56915
    },
    {
      "entropy": 5.564244079589844,
      "epoch": 5.163280116110305,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002665634238562052,
      "loss": 4.9678,
      "mean_token_accuracy": 0.22466486543416977,
      "num_tokens": 118118481.0,
      "step": 56920
    },
    {
      "entropy": 5.603254795074463,
      "epoch": 5.163733671988389,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002665310902773733,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22642994821071624,
      "num_tokens": 118128236.0,
      "step": 56925
    },
    {
      "entropy": 5.536253118515015,
      "epoch": 5.164187227866473,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00026649875687368053,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2291414961218834,
      "num_tokens": 118138967.0,
      "step": 56930
    },
    {
      "entropy": 5.550162410736084,
      "epoch": 5.164640783744558,
      "grad_norm": 1.125,
      "learning_rate": 0.00026646642364579567,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21718933284282685,
      "num_tokens": 118148920.0,
      "step": 56935
    },
    {
      "entropy": 5.496624708175659,
      "epoch": 5.165094339622642,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026643409059438726,
      "loss": 4.824,
      "mean_token_accuracy": 0.22607936412096025,
      "num_tokens": 118158184.0,
      "step": 56940
    },
    {
      "entropy": 5.5300037384033205,
      "epoch": 5.165547895500726,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026640175772012415,
      "loss": 4.838,
      "mean_token_accuracy": 0.22616789042949675,
      "num_tokens": 118168785.0,
      "step": 56945
    },
    {
      "entropy": 5.504164218902588,
      "epoch": 5.1660014513788095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026636942502367486,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21727697551250458,
      "num_tokens": 118179713.0,
      "step": 56950
    },
    {
      "entropy": 5.485883140563965,
      "epoch": 5.166455007256894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000266337092505708,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22633203417062758,
      "num_tokens": 118190226.0,
      "step": 56955
    },
    {
      "entropy": 5.482411289215088,
      "epoch": 5.166908563134978,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002663047601668923,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2314671441912651,
      "num_tokens": 118200649.0,
      "step": 56960
    },
    {
      "entropy": 5.516838598251343,
      "epoch": 5.167362119013062,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002662724280078964,
      "loss": 4.9394,
      "mean_token_accuracy": 0.22605040073394775,
      "num_tokens": 118211249.0,
      "step": 56965
    },
    {
      "entropy": 5.565868091583252,
      "epoch": 5.167815674891147,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002662400960293889,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2225196674466133,
      "num_tokens": 118221734.0,
      "step": 56970
    },
    {
      "entropy": 5.516733884811401,
      "epoch": 5.168269230769231,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026620776423203846,
      "loss": 4.8704,
      "mean_token_accuracy": 0.22981493920087814,
      "num_tokens": 118232371.0,
      "step": 56975
    },
    {
      "entropy": 5.545538902282715,
      "epoch": 5.168722786647315,
      "grad_norm": 1.125,
      "learning_rate": 0.00026617543261651374,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22624792605638505,
      "num_tokens": 118242830.0,
      "step": 56980
    },
    {
      "entropy": 5.462700510025025,
      "epoch": 5.169176342525399,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002661431011834833,
      "loss": 4.7845,
      "mean_token_accuracy": 0.23251079469919206,
      "num_tokens": 118253415.0,
      "step": 56985
    },
    {
      "entropy": 5.524801349639892,
      "epoch": 5.169629898403484,
      "grad_norm": 1.125,
      "learning_rate": 0.00026611076993361574,
      "loss": 4.9,
      "mean_token_accuracy": 0.22440169751644135,
      "num_tokens": 118263744.0,
      "step": 56990
    },
    {
      "entropy": 5.482697916030884,
      "epoch": 5.170083454281568,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026607843886757964,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22426964938640595,
      "num_tokens": 118274563.0,
      "step": 56995
    },
    {
      "entropy": 5.534286212921143,
      "epoch": 5.170537010159651,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026604610798604386,
      "loss": 4.9807,
      "mean_token_accuracy": 0.22386185675859452,
      "num_tokens": 118285152.0,
      "step": 57000
    },
    {
      "epoch": 5.170537010159651,
      "eval_entropy": 5.217111354931447,
      "eval_loss": 4.994021892547607,
      "eval_mean_token_accuracy": 0.22825295515864144,
      "eval_num_tokens": 118285152.0,
      "eval_runtime": 26.6822,
      "eval_samples_per_second": 1325.227,
      "eval_steps_per_second": 165.653,
      "step": 57000
    }
  ],
  "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": 1.73112573416832e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}