{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0811249323958896,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.7960851192474365,
      "epoch": 0.005408328826392645,
      "grad_norm": 27.25,
      "learning_rate": 2e-06,
      "loss": 14.2304,
      "mean_token_accuracy": 0.0003160792635753751,
      "num_tokens": 10942.0,
      "step": 5
    },
    {
      "entropy": 4.8001548767089846,
      "epoch": 0.01081665765278529,
      "grad_norm": 34.0,
      "learning_rate": 4.5e-06,
      "loss": 14.2056,
      "mean_token_accuracy": 0.0001242236001417041,
      "num_tokens": 21887.0,
      "step": 10
    },
    {
      "entropy": 5.043719053268433,
      "epoch": 0.016224986479177934,
      "grad_norm": 36.5,
      "learning_rate": 7e-06,
      "loss": 14.0392,
      "mean_token_accuracy": 6.711409660056233e-05,
      "num_tokens": 34345.0,
      "step": 15
    },
    {
      "entropy": 5.49084243774414,
      "epoch": 0.02163331530557058,
      "grad_norm": 34.75,
      "learning_rate": 9.5e-06,
      "loss": 13.6544,
      "mean_token_accuracy": 9.285050909966231e-05,
      "num_tokens": 45445.0,
      "step": 20
    },
    {
      "entropy": 7.0108997344970705,
      "epoch": 0.02704164413196322,
      "grad_norm": 19.25,
      "learning_rate": 1.2e-05,
      "loss": 12.8774,
      "mean_token_accuracy": 0.00023736039875075222,
      "num_tokens": 58237.0,
      "step": 25
    },
    {
      "entropy": 8.019237947463989,
      "epoch": 0.03244997295835587,
      "grad_norm": 9.3125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 12.3405,
      "mean_token_accuracy": 9.813542710617185e-05,
      "num_tokens": 68098.0,
      "step": 30
    },
    {
      "entropy": 9.054293632507324,
      "epoch": 0.03785830178474851,
      "grad_norm": 5.46875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 11.9463,
      "mean_token_accuracy": 0.0,
      "num_tokens": 79867.0,
      "step": 35
    },
    {
      "entropy": 9.511068725585938,
      "epoch": 0.04326663061114116,
      "grad_norm": 3.796875,
      "learning_rate": 1.95e-05,
      "loss": 11.6959,
      "mean_token_accuracy": 0.0,
      "num_tokens": 91257.0,
      "step": 40
    },
    {
      "entropy": 9.867530632019044,
      "epoch": 0.048674959437533805,
      "grad_norm": 5.53125,
      "learning_rate": 2.2e-05,
      "loss": 11.45,
      "mean_token_accuracy": 9.38967161346227e-05,
      "num_tokens": 103474.0,
      "step": 45
    },
    {
      "entropy": 10.309116744995118,
      "epoch": 0.05408328826392644,
      "grad_norm": 4.5,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 11.0292,
      "mean_token_accuracy": 0.0003331616928335279,
      "num_tokens": 114830.0,
      "step": 50
    },
    {
      "entropy": 10.523942279815675,
      "epoch": 0.05949161709031909,
      "grad_norm": 3.921875,
      "learning_rate": 2.7e-05,
      "loss": 10.774,
      "mean_token_accuracy": 0.011405656964052468,
      "num_tokens": 126017.0,
      "step": 55
    },
    {
      "entropy": 10.591793823242188,
      "epoch": 0.06489994591671173,
      "grad_norm": 2.734375,
      "learning_rate": 2.95e-05,
      "loss": 10.5283,
      "mean_token_accuracy": 0.03312199003994465,
      "num_tokens": 138509.0,
      "step": 60
    },
    {
      "entropy": 10.653960514068604,
      "epoch": 0.07030827474310439,
      "grad_norm": 2.375,
      "learning_rate": 3.2e-05,
      "loss": 10.3296,
      "mean_token_accuracy": 0.032539681531488895,
      "num_tokens": 148353.0,
      "step": 65
    },
    {
      "entropy": 10.66145372390747,
      "epoch": 0.07571660356949703,
      "grad_norm": 2.140625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 10.2093,
      "mean_token_accuracy": 0.031699617952108385,
      "num_tokens": 160269.0,
      "step": 70
    },
    {
      "entropy": 10.610020542144776,
      "epoch": 0.08112493239588967,
      "grad_norm": 2.0,
      "learning_rate": 3.7e-05,
      "loss": 10.0327,
      "mean_token_accuracy": 0.03554317634552717,
      "num_tokens": 171568.0,
      "step": 75
    },
    {
      "entropy": 10.5712140083313,
      "epoch": 0.08653326122228232,
      "grad_norm": 1.984375,
      "learning_rate": 3.95e-05,
      "loss": 9.8889,
      "mean_token_accuracy": 0.03888488262891769,
      "num_tokens": 184243.0,
      "step": 80
    },
    {
      "entropy": 10.565914058685303,
      "epoch": 0.09194159004867496,
      "grad_norm": 1.96875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.7667,
      "mean_token_accuracy": 0.03745253346860409,
      "num_tokens": 194887.0,
      "step": 85
    },
    {
      "entropy": 10.57132396697998,
      "epoch": 0.09734991887506761,
      "grad_norm": 1.9296875,
      "learning_rate": 4.45e-05,
      "loss": 9.6353,
      "mean_token_accuracy": 0.03890889976173639,
      "num_tokens": 206476.0,
      "step": 90
    },
    {
      "entropy": 10.575814723968506,
      "epoch": 0.10275824770146025,
      "grad_norm": 1.8671875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4592,
      "mean_token_accuracy": 0.03984561040997505,
      "num_tokens": 216578.0,
      "step": 95
    },
    {
      "entropy": 10.575993537902832,
      "epoch": 0.10816657652785289,
      "grad_norm": 1.8515625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3575,
      "mean_token_accuracy": 0.04156542681157589,
      "num_tokens": 226941.0,
      "step": 100
    },
    {
      "entropy": 10.572124099731445,
      "epoch": 0.11357490535424554,
      "grad_norm": 1.828125,
      "learning_rate": 5.2e-05,
      "loss": 9.2249,
      "mean_token_accuracy": 0.03909948579967022,
      "num_tokens": 238060.0,
      "step": 105
    },
    {
      "entropy": 10.561905097961425,
      "epoch": 0.11898323418063818,
      "grad_norm": 1.9453125,
      "learning_rate": 5.45e-05,
      "loss": 9.0694,
      "mean_token_accuracy": 0.04209643751382828,
      "num_tokens": 249137.0,
      "step": 110
    },
    {
      "entropy": 10.530157947540284,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.828125,
      "learning_rate": 5.7e-05,
      "loss": 8.9517,
      "mean_token_accuracy": 0.042229291051626205,
      "num_tokens": 260712.0,
      "step": 115
    },
    {
      "entropy": 10.49021110534668,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.84375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.7898,
      "mean_token_accuracy": 0.0394113190472126,
      "num_tokens": 271039.0,
      "step": 120
    },
    {
      "entropy": 10.455241107940674,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.921875,
      "learning_rate": 6.2e-05,
      "loss": 8.6467,
      "mean_token_accuracy": 0.040281732007861136,
      "num_tokens": 281353.0,
      "step": 125
    },
    {
      "entropy": 10.391654300689698,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.6015625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.5617,
      "mean_token_accuracy": 0.04575536660850048,
      "num_tokens": 292221.0,
      "step": 130
    },
    {
      "entropy": 10.302424240112305,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.5625,
      "learning_rate": 6.7e-05,
      "loss": 8.4303,
      "mean_token_accuracy": 0.0425492487847805,
      "num_tokens": 303353.0,
      "step": 135
    },
    {
      "entropy": 10.208961296081544,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.4453125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3137,
      "mean_token_accuracy": 0.04581625089049339,
      "num_tokens": 314165.0,
      "step": 140
    },
    {
      "entropy": 10.080995559692383,
      "epoch": 0.1568415359653867,
      "grad_norm": 1.40625,
      "learning_rate": 7.2e-05,
      "loss": 8.1479,
      "mean_token_accuracy": 0.049863246828317644,
      "num_tokens": 326387.0,
      "step": 145
    },
    {
      "entropy": 9.93353729248047,
      "epoch": 0.16224986479177933,
      "grad_norm": 1.3359375,
      "learning_rate": 7.45e-05,
      "loss": 8.0989,
      "mean_token_accuracy": 0.04789762534201145,
      "num_tokens": 339155.0,
      "step": 150
    },
    {
      "entropy": 9.778404998779298,
      "epoch": 0.167658193618172,
      "grad_norm": 1.359375,
      "learning_rate": 7.7e-05,
      "loss": 7.9369,
      "mean_token_accuracy": 0.04837009124457836,
      "num_tokens": 350643.0,
      "step": 155
    },
    {
      "entropy": 9.568048095703125,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.171875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.8363,
      "mean_token_accuracy": 0.04552422165870666,
      "num_tokens": 361977.0,
      "step": 160
    },
    {
      "entropy": 9.345962619781494,
      "epoch": 0.17847485127095727,
      "grad_norm": 0.9609375,
      "learning_rate": 8.2e-05,
      "loss": 7.8056,
      "mean_token_accuracy": 0.048871561884880066,
      "num_tokens": 373604.0,
      "step": 165
    },
    {
      "entropy": 9.073436737060547,
      "epoch": 0.1838831800973499,
      "grad_norm": 0.83203125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7117,
      "mean_token_accuracy": 0.04806899577379227,
      "num_tokens": 384306.0,
      "step": 170
    },
    {
      "entropy": 8.822536659240722,
      "epoch": 0.18929150892374255,
      "grad_norm": 0.77734375,
      "learning_rate": 8.7e-05,
      "loss": 7.5951,
      "mean_token_accuracy": 0.04366670176386833,
      "num_tokens": 396479.0,
      "step": 175
    },
    {
      "entropy": 8.54340581893921,
      "epoch": 0.19469983775013522,
      "grad_norm": 0.74609375,
      "learning_rate": 8.95e-05,
      "loss": 7.5166,
      "mean_token_accuracy": 0.05379514247179031,
      "num_tokens": 408489.0,
      "step": 180
    },
    {
      "entropy": 8.377008247375489,
      "epoch": 0.20010816657652786,
      "grad_norm": 0.671875,
      "learning_rate": 9.2e-05,
      "loss": 7.5106,
      "mean_token_accuracy": 0.06423841267824174,
      "num_tokens": 419339.0,
      "step": 185
    },
    {
      "entropy": 8.208265113830567,
      "epoch": 0.2055164954029205,
      "grad_norm": 0.67578125,
      "learning_rate": 9.45e-05,
      "loss": 7.4712,
      "mean_token_accuracy": 0.06963393315672875,
      "num_tokens": 430486.0,
      "step": 190
    },
    {
      "entropy": 8.0866042137146,
      "epoch": 0.21092482422931313,
      "grad_norm": 0.75,
      "learning_rate": 9.7e-05,
      "loss": 7.4643,
      "mean_token_accuracy": 0.06951842345297336,
      "num_tokens": 440414.0,
      "step": 195
    },
    {
      "entropy": 7.999132299423218,
      "epoch": 0.21633315305570577,
      "grad_norm": 0.6640625,
      "learning_rate": 9.95e-05,
      "loss": 7.4064,
      "mean_token_accuracy": 0.07127611935138703,
      "num_tokens": 451475.0,
      "step": 200
    },
    {
      "entropy": 7.92455677986145,
      "epoch": 0.22174148188209844,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000102,
      "loss": 7.4544,
      "mean_token_accuracy": 0.0808640293776989,
      "num_tokens": 462053.0,
      "step": 205
    },
    {
      "entropy": 7.913168287277221,
      "epoch": 0.22714981070849108,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4165,
      "mean_token_accuracy": 0.07268866151571274,
      "num_tokens": 473207.0,
      "step": 210
    },
    {
      "entropy": 7.920434427261353,
      "epoch": 0.23255813953488372,
      "grad_norm": 0.75,
      "learning_rate": 0.000107,
      "loss": 7.3957,
      "mean_token_accuracy": 0.07367360889911652,
      "num_tokens": 484999.0,
      "step": 215
    },
    {
      "entropy": 7.8919206142425535,
      "epoch": 0.23796646836127636,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0001095,
      "loss": 7.3913,
      "mean_token_accuracy": 0.07248864844441413,
      "num_tokens": 496772.0,
      "step": 220
    },
    {
      "entropy": 7.9051414966583256,
      "epoch": 0.24337479718766902,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000112,
      "loss": 7.3555,
      "mean_token_accuracy": 0.07366742193698883,
      "num_tokens": 507782.0,
      "step": 225
    },
    {
      "entropy": 7.845500946044922,
      "epoch": 0.24878312601406166,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0001145,
      "loss": 7.2834,
      "mean_token_accuracy": 0.07393435426056386,
      "num_tokens": 518950.0,
      "step": 230
    },
    {
      "entropy": 7.782898235321045,
      "epoch": 0.2541914548404543,
      "grad_norm": 0.796875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3481,
      "mean_token_accuracy": 0.07736928686499596,
      "num_tokens": 530795.0,
      "step": 235
    },
    {
      "entropy": 7.807211065292359,
      "epoch": 0.25959978366684694,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3554,
      "mean_token_accuracy": 0.07271409779787064,
      "num_tokens": 542878.0,
      "step": 240
    },
    {
      "entropy": 7.8081248760223385,
      "epoch": 0.2650081124932396,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000122,
      "loss": 7.2632,
      "mean_token_accuracy": 0.07816940769553185,
      "num_tokens": 553682.0,
      "step": 245
    },
    {
      "entropy": 7.717547512054443,
      "epoch": 0.2704164413196322,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001245,
      "loss": 7.2621,
      "mean_token_accuracy": 0.08631859943270684,
      "num_tokens": 564433.0,
      "step": 250
    },
    {
      "entropy": 7.729435443878174,
      "epoch": 0.27582477014602486,
      "grad_norm": 0.7109375,
      "learning_rate": 0.000127,
      "loss": 7.28,
      "mean_token_accuracy": 0.0831921637058258,
      "num_tokens": 576001.0,
      "step": 255
    },
    {
      "entropy": 7.762239027023315,
      "epoch": 0.28123309897241755,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001295,
      "loss": 7.2116,
      "mean_token_accuracy": 0.08321145027875901,
      "num_tokens": 587138.0,
      "step": 260
    },
    {
      "entropy": 7.712756729125976,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.03125,
      "learning_rate": 0.000132,
      "loss": 7.1655,
      "mean_token_accuracy": 0.08543640077114105,
      "num_tokens": 597572.0,
      "step": 265
    },
    {
      "entropy": 7.693479919433594,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.015625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.1296,
      "mean_token_accuracy": 0.09074783772230148,
      "num_tokens": 608879.0,
      "step": 270
    },
    {
      "entropy": 7.604578304290771,
      "epoch": 0.29745808545159547,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1746,
      "mean_token_accuracy": 0.08316268846392631,
      "num_tokens": 620439.0,
      "step": 275
    },
    {
      "entropy": 7.675649356842041,
      "epoch": 0.3028664142779881,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0001395,
      "loss": 7.2274,
      "mean_token_accuracy": 0.0821447916328907,
      "num_tokens": 632543.0,
      "step": 280
    },
    {
      "entropy": 7.693741321563721,
      "epoch": 0.30827474310438074,
      "grad_norm": 0.75,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.258,
      "mean_token_accuracy": 0.09081905633211136,
      "num_tokens": 643369.0,
      "step": 285
    },
    {
      "entropy": 7.64626145362854,
      "epoch": 0.3136830719307734,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0001445,
      "loss": 7.1271,
      "mean_token_accuracy": 0.08910127580165864,
      "num_tokens": 655328.0,
      "step": 290
    },
    {
      "entropy": 7.68002290725708,
      "epoch": 0.319091400757166,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000147,
      "loss": 7.1774,
      "mean_token_accuracy": 0.08911841437220573,
      "num_tokens": 665780.0,
      "step": 295
    },
    {
      "entropy": 7.583356475830078,
      "epoch": 0.32449972958355866,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0001495,
      "loss": 7.0876,
      "mean_token_accuracy": 0.09305166527628898,
      "num_tokens": 676566.0,
      "step": 300
    },
    {
      "entropy": 7.611654996871948,
      "epoch": 0.3299080584099513,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000152,
      "loss": 7.0847,
      "mean_token_accuracy": 0.09290368705987931,
      "num_tokens": 687407.0,
      "step": 305
    },
    {
      "entropy": 7.593765163421631,
      "epoch": 0.335316387236344,
      "grad_norm": 1.015625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0044,
      "mean_token_accuracy": 0.09596289992332459,
      "num_tokens": 698162.0,
      "step": 310
    },
    {
      "entropy": 7.452439403533935,
      "epoch": 0.34072471606273663,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000157,
      "loss": 7.0018,
      "mean_token_accuracy": 0.08815811201930046,
      "num_tokens": 708743.0,
      "step": 315
    },
    {
      "entropy": 7.615065431594848,
      "epoch": 0.34613304488912927,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0001595,
      "loss": 7.1876,
      "mean_token_accuracy": 0.09327791407704353,
      "num_tokens": 719828.0,
      "step": 320
    },
    {
      "entropy": 7.63629150390625,
      "epoch": 0.3515413737155219,
      "grad_norm": 0.90625,
      "learning_rate": 0.000162,
      "loss": 7.149,
      "mean_token_accuracy": 0.08833655528724194,
      "num_tokens": 732514.0,
      "step": 325
    },
    {
      "entropy": 7.527328824996948,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.9889,
      "mean_token_accuracy": 0.09440577700734139,
      "num_tokens": 742391.0,
      "step": 330
    },
    {
      "entropy": 7.36641492843628,
      "epoch": 0.3623580313683072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9959,
      "mean_token_accuracy": 0.08885396271944046,
      "num_tokens": 753638.0,
      "step": 335
    },
    {
      "entropy": 7.65393443107605,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.1896,
      "mean_token_accuracy": 0.09420613124966622,
      "num_tokens": 764850.0,
      "step": 340
    },
    {
      "entropy": 7.468115282058716,
      "epoch": 0.37317468902109246,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9635,
      "mean_token_accuracy": 0.10290685519576073,
      "num_tokens": 776192.0,
      "step": 345
    },
    {
      "entropy": 7.411421632766723,
      "epoch": 0.3785830178474851,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9167,
      "mean_token_accuracy": 0.10349498242139817,
      "num_tokens": 787051.0,
      "step": 350
    },
    {
      "entropy": 7.389636278152466,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.015625,
      "learning_rate": 0.000177,
      "loss": 6.9474,
      "mean_token_accuracy": 0.09752821102738381,
      "num_tokens": 799281.0,
      "step": 355
    },
    {
      "entropy": 7.4360168933868405,
      "epoch": 0.38939967550027044,
      "grad_norm": 0.84375,
      "learning_rate": 0.0001795,
      "loss": 6.9509,
      "mean_token_accuracy": 0.09831701442599297,
      "num_tokens": 810003.0,
      "step": 360
    },
    {
      "entropy": 7.410728406906128,
      "epoch": 0.3948080043266631,
      "grad_norm": 0.96875,
      "learning_rate": 0.000182,
      "loss": 6.9213,
      "mean_token_accuracy": 0.09893173575401307,
      "num_tokens": 819749.0,
      "step": 365
    },
    {
      "entropy": 7.420523643493652,
      "epoch": 0.4002163331530557,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0001845,
      "loss": 6.9334,
      "mean_token_accuracy": 0.10455422475934029,
      "num_tokens": 830552.0,
      "step": 370
    },
    {
      "entropy": 7.477943468093872,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.3125,
      "learning_rate": 0.000187,
      "loss": 7.0213,
      "mean_token_accuracy": 0.10054472163319587,
      "num_tokens": 840495.0,
      "step": 375
    },
    {
      "entropy": 7.325655174255371,
      "epoch": 0.411032990805841,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0001895,
      "loss": 6.9394,
      "mean_token_accuracy": 0.10164839625358582,
      "num_tokens": 852915.0,
      "step": 380
    },
    {
      "entropy": 7.386038827896118,
      "epoch": 0.41644131963223363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000192,
      "loss": 6.9133,
      "mean_token_accuracy": 0.10262698084115982,
      "num_tokens": 862762.0,
      "step": 385
    },
    {
      "entropy": 7.399918270111084,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001945,
      "loss": 6.8485,
      "mean_token_accuracy": 0.10503640621900559,
      "num_tokens": 873023.0,
      "step": 390
    },
    {
      "entropy": 7.350394630432129,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.9557,
      "mean_token_accuracy": 0.0971641018986702,
      "num_tokens": 885281.0,
      "step": 395
    },
    {
      "entropy": 7.2543981075286865,
      "epoch": 0.43266630611141155,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9465,
      "mean_token_accuracy": 0.10591788291931152,
      "num_tokens": 896210.0,
      "step": 400
    },
    {
      "entropy": 7.301179552078247,
      "epoch": 0.43807463493780424,
      "grad_norm": 0.78515625,
      "learning_rate": 0.000202,
      "loss": 6.8552,
      "mean_token_accuracy": 0.10582882538437843,
      "num_tokens": 909028.0,
      "step": 405
    },
    {
      "entropy": 7.2928071975708,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.0,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7897,
      "mean_token_accuracy": 0.10721533969044686,
      "num_tokens": 919710.0,
      "step": 410
    },
    {
      "entropy": 7.37521538734436,
      "epoch": 0.4488912925905895,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000207,
      "loss": 6.8504,
      "mean_token_accuracy": 0.10534188896417618,
      "num_tokens": 930019.0,
      "step": 415
    },
    {
      "entropy": 7.277246475219727,
      "epoch": 0.45429962141698216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002095,
      "loss": 6.7672,
      "mean_token_accuracy": 0.11057652086019516,
      "num_tokens": 940535.0,
      "step": 420
    },
    {
      "entropy": 7.120538902282715,
      "epoch": 0.4597079502433748,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000212,
      "loss": 6.7261,
      "mean_token_accuracy": 0.10942267924547196,
      "num_tokens": 951390.0,
      "step": 425
    },
    {
      "entropy": 7.226644659042359,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002145,
      "loss": 6.7662,
      "mean_token_accuracy": 0.11120115816593171,
      "num_tokens": 963010.0,
      "step": 430
    },
    {
      "entropy": 7.179320001602173,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.7145,
      "mean_token_accuracy": 0.11507150083780289,
      "num_tokens": 974032.0,
      "step": 435
    },
    {
      "entropy": 7.230346393585205,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002195,
      "loss": 6.7698,
      "mean_token_accuracy": 0.11191772520542145,
      "num_tokens": 983954.0,
      "step": 440
    },
    {
      "entropy": 7.1874768257141115,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.109375,
      "learning_rate": 0.000222,
      "loss": 6.7844,
      "mean_token_accuracy": 0.1129900112748146,
      "num_tokens": 994135.0,
      "step": 445
    },
    {
      "entropy": 7.209841632843018,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002245,
      "loss": 6.7424,
      "mean_token_accuracy": 0.10967175662517548,
      "num_tokens": 1006275.0,
      "step": 450
    },
    {
      "entropy": 7.0370276927947994,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.5695,
      "mean_token_accuracy": 0.11895088031888008,
      "num_tokens": 1017767.0,
      "step": 455
    },
    {
      "entropy": 7.171165800094604,
      "epoch": 0.4975662520281233,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6998,
      "mean_token_accuracy": 0.11892824321985244,
      "num_tokens": 1028286.0,
      "step": 460
    },
    {
      "entropy": 7.08244252204895,
      "epoch": 0.502974580854516,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.71,
      "mean_token_accuracy": 0.11281011700630188,
      "num_tokens": 1040555.0,
      "step": 465
    },
    {
      "entropy": 7.177019929885864,
      "epoch": 0.5083829096809086,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8046,
      "mean_token_accuracy": 0.11459487229585648,
      "num_tokens": 1053668.0,
      "step": 470
    },
    {
      "entropy": 7.1339305400848385,
      "epoch": 0.5137912385073012,
      "grad_norm": 0.96875,
      "learning_rate": 0.000237,
      "loss": 6.7437,
      "mean_token_accuracy": 0.11368778720498085,
      "num_tokens": 1066042.0,
      "step": 475
    },
    {
      "entropy": 7.061350917816162,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002395,
      "loss": 6.7603,
      "mean_token_accuracy": 0.11287353932857513,
      "num_tokens": 1077843.0,
      "step": 480
    },
    {
      "entropy": 7.126706933975219,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000242,
      "loss": 6.6459,
      "mean_token_accuracy": 0.11803325861692429,
      "num_tokens": 1089448.0,
      "step": 485
    },
    {
      "entropy": 7.057846260070801,
      "epoch": 0.5300162249864792,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002445,
      "loss": 6.7488,
      "mean_token_accuracy": 0.11246698349714279,
      "num_tokens": 1100989.0,
      "step": 490
    },
    {
      "entropy": 7.211519956588745,
      "epoch": 0.5354245538128718,
      "grad_norm": 0.890625,
      "learning_rate": 0.000247,
      "loss": 6.7773,
      "mean_token_accuracy": 0.11612514406442642,
      "num_tokens": 1113012.0,
      "step": 495
    },
    {
      "entropy": 7.058304357528686,
      "epoch": 0.5408328826392644,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002495,
      "loss": 6.7171,
      "mean_token_accuracy": 0.11422862857580185,
      "num_tokens": 1123551.0,
      "step": 500
    },
    {
      "epoch": 0.5408328826392644,
      "eval_entropy": 7.070621424371546,
      "eval_loss": 6.859935283660889,
      "eval_mean_token_accuracy": 0.11617732908238064,
      "eval_num_tokens": 1123551.0,
      "eval_runtime": 1.9446,
      "eval_samples_per_second": 1807.585,
      "eval_steps_per_second": 226.27,
      "step": 500
    },
    {
      "entropy": 7.120835399627685,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000252,
      "loss": 6.6374,
      "mean_token_accuracy": 0.11516082882881165,
      "num_tokens": 1133838.0,
      "step": 505
    },
    {
      "entropy": 7.017620658874511,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002545,
      "loss": 6.6809,
      "mean_token_accuracy": 0.1152262344956398,
      "num_tokens": 1146005.0,
      "step": 510
    },
    {
      "entropy": 7.052067136764526,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.046875,
      "learning_rate": 0.000257,
      "loss": 6.7621,
      "mean_token_accuracy": 0.11798961162567138,
      "num_tokens": 1157328.0,
      "step": 515
    },
    {
      "entropy": 6.988781690597534,
      "epoch": 0.5624661979448351,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002595,
      "loss": 6.5722,
      "mean_token_accuracy": 0.12257469147443771,
      "num_tokens": 1168592.0,
      "step": 520
    },
    {
      "entropy": 7.097140264511109,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.09375,
      "learning_rate": 0.000262,
      "loss": 6.6913,
      "mean_token_accuracy": 0.11780922934412956,
      "num_tokens": 1178680.0,
      "step": 525
    },
    {
      "entropy": 7.017797470092773,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6329,
      "mean_token_accuracy": 0.12091006338596344,
      "num_tokens": 1188825.0,
      "step": 530
    },
    {
      "entropy": 7.020737600326538,
      "epoch": 0.578691184424013,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6649,
      "mean_token_accuracy": 0.121755550801754,
      "num_tokens": 1201375.0,
      "step": 535
    },
    {
      "entropy": 7.082882022857666,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7857,
      "mean_token_accuracy": 0.11599067747592925,
      "num_tokens": 1211915.0,
      "step": 540
    },
    {
      "entropy": 7.0102982997894285,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7015,
      "mean_token_accuracy": 0.12261592373251914,
      "num_tokens": 1223151.0,
      "step": 545
    },
    {
      "entropy": 7.062853527069092,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002745,
      "loss": 6.6242,
      "mean_token_accuracy": 0.12241948544979095,
      "num_tokens": 1234229.0,
      "step": 550
    },
    {
      "entropy": 7.009330701828003,
      "epoch": 0.6003244997295836,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000277,
      "loss": 6.6335,
      "mean_token_accuracy": 0.12372676730155945,
      "num_tokens": 1245298.0,
      "step": 555
    },
    {
      "entropy": 6.956117057800293,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002795,
      "loss": 6.6296,
      "mean_token_accuracy": 0.12371465265750885,
      "num_tokens": 1256154.0,
      "step": 560
    },
    {
      "entropy": 6.9070703983306885,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5398,
      "mean_token_accuracy": 0.12442566901445389,
      "num_tokens": 1267404.0,
      "step": 565
    },
    {
      "entropy": 6.95917820930481,
      "epoch": 0.6165494862087615,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002845,
      "loss": 6.5921,
      "mean_token_accuracy": 0.12321074232459069,
      "num_tokens": 1278894.0,
      "step": 570
    },
    {
      "entropy": 6.912023448944092,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000287,
      "loss": 6.5839,
      "mean_token_accuracy": 0.12680861726403236,
      "num_tokens": 1290408.0,
      "step": 575
    },
    {
      "entropy": 7.008191251754761,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002895,
      "loss": 6.6473,
      "mean_token_accuracy": 0.12115282118320465,
      "num_tokens": 1301508.0,
      "step": 580
    },
    {
      "entropy": 7.004787969589233,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.03125,
      "learning_rate": 0.000292,
      "loss": 6.624,
      "mean_token_accuracy": 0.12800585329532624,
      "num_tokens": 1312551.0,
      "step": 585
    },
    {
      "entropy": 6.99333176612854,
      "epoch": 0.638182801514332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002945,
      "loss": 6.7226,
      "mean_token_accuracy": 0.12242905274033547,
      "num_tokens": 1324876.0,
      "step": 590
    },
    {
      "entropy": 7.026355409622193,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.03125,
      "learning_rate": 0.000297,
      "loss": 6.6088,
      "mean_token_accuracy": 0.12180419266223907,
      "num_tokens": 1335401.0,
      "step": 595
    },
    {
      "entropy": 6.767445611953735,
      "epoch": 0.6489994591671173,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002995,
      "loss": 6.538,
      "mean_token_accuracy": 0.12269261777400971,
      "num_tokens": 1348265.0,
      "step": 600
    },
    {
      "entropy": 6.939475727081299,
      "epoch": 0.65440778799351,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000302,
      "loss": 6.4984,
      "mean_token_accuracy": 0.12996519654989241,
      "num_tokens": 1359078.0,
      "step": 605
    },
    {
      "entropy": 6.676696491241455,
      "epoch": 0.6598161168199026,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003045,
      "loss": 6.3855,
      "mean_token_accuracy": 0.12842316627502443,
      "num_tokens": 1369499.0,
      "step": 610
    },
    {
      "entropy": 6.887877798080444,
      "epoch": 0.6652244456462953,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000307,
      "loss": 6.4993,
      "mean_token_accuracy": 0.12684686109423637,
      "num_tokens": 1380329.0,
      "step": 615
    },
    {
      "entropy": 6.855057859420777,
      "epoch": 0.670632774472688,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0003095,
      "loss": 6.5755,
      "mean_token_accuracy": 0.12310040593147278,
      "num_tokens": 1393047.0,
      "step": 620
    },
    {
      "entropy": 6.847856616973877,
      "epoch": 0.6760411032990806,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000312,
      "loss": 6.5675,
      "mean_token_accuracy": 0.12307048961520195,
      "num_tokens": 1405600.0,
      "step": 625
    },
    {
      "entropy": 6.890627050399781,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003145,
      "loss": 6.5113,
      "mean_token_accuracy": 0.12754200771450996,
      "num_tokens": 1416670.0,
      "step": 630
    },
    {
      "entropy": 6.845398712158203,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.03125,
      "learning_rate": 0.000317,
      "loss": 6.5305,
      "mean_token_accuracy": 0.12670771926641464,
      "num_tokens": 1428929.0,
      "step": 635
    },
    {
      "entropy": 6.809947156906128,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003195,
      "loss": 6.4476,
      "mean_token_accuracy": 0.12762413546442986,
      "num_tokens": 1439298.0,
      "step": 640
    },
    {
      "entropy": 6.87408938407898,
      "epoch": 0.6976744186046512,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000322,
      "loss": 6.5514,
      "mean_token_accuracy": 0.13081721141934394,
      "num_tokens": 1449853.0,
      "step": 645
    },
    {
      "entropy": 6.944610214233398,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6146,
      "mean_token_accuracy": 0.12372421324253083,
      "num_tokens": 1460816.0,
      "step": 650
    },
    {
      "entropy": 6.768828868865967,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4671,
      "mean_token_accuracy": 0.12459458410739899,
      "num_tokens": 1473003.0,
      "step": 655
    },
    {
      "entropy": 6.815493726730347,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5935,
      "mean_token_accuracy": 0.12046970501542091,
      "num_tokens": 1484745.0,
      "step": 660
    },
    {
      "entropy": 6.918433952331543,
      "epoch": 0.7193077339102217,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4568,
      "mean_token_accuracy": 0.1293526917695999,
      "num_tokens": 1495803.0,
      "step": 665
    },
    {
      "entropy": 6.756926679611206,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4795,
      "mean_token_accuracy": 0.13275134861469268,
      "num_tokens": 1507504.0,
      "step": 670
    },
    {
      "entropy": 6.846708583831787,
      "epoch": 0.730124391563007,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000337,
      "loss": 6.5749,
      "mean_token_accuracy": 0.12511131837964057,
      "num_tokens": 1518765.0,
      "step": 675
    },
    {
      "entropy": 6.733471632003784,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003395,
      "loss": 6.4555,
      "mean_token_accuracy": 0.12526600286364556,
      "num_tokens": 1529481.0,
      "step": 680
    },
    {
      "entropy": 6.803652429580689,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000342,
      "loss": 6.4681,
      "mean_token_accuracy": 0.1339200973510742,
      "num_tokens": 1539612.0,
      "step": 685
    },
    {
      "entropy": 6.806907892227173,
      "epoch": 0.7463493780421849,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6046,
      "mean_token_accuracy": 0.12426809221506119,
      "num_tokens": 1551370.0,
      "step": 690
    },
    {
      "entropy": 6.894266033172608,
      "epoch": 0.7517577068685776,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000347,
      "loss": 6.4958,
      "mean_token_accuracy": 0.13355423510074615,
      "num_tokens": 1561662.0,
      "step": 695
    },
    {
      "entropy": 6.808643198013305,
      "epoch": 0.7571660356949702,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003495,
      "loss": 6.4794,
      "mean_token_accuracy": 0.12898182794451712,
      "num_tokens": 1573479.0,
      "step": 700
    },
    {
      "entropy": 6.713399219512939,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000352,
      "loss": 6.4255,
      "mean_token_accuracy": 0.13072630763053894,
      "num_tokens": 1583684.0,
      "step": 705
    },
    {
      "entropy": 6.775609922409058,
      "epoch": 0.7679826933477556,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003545,
      "loss": 6.4727,
      "mean_token_accuracy": 0.13058030232787132,
      "num_tokens": 1595694.0,
      "step": 710
    },
    {
      "entropy": 6.718004512786865,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.125,
      "learning_rate": 0.000357,
      "loss": 6.4717,
      "mean_token_accuracy": 0.12594751939177512,
      "num_tokens": 1606104.0,
      "step": 715
    },
    {
      "entropy": 6.8197935104370115,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003595,
      "loss": 6.541,
      "mean_token_accuracy": 0.1298747517168522,
      "num_tokens": 1617368.0,
      "step": 720
    },
    {
      "entropy": 6.853786754608154,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000362,
      "loss": 6.5568,
      "mean_token_accuracy": 0.12204690799117088,
      "num_tokens": 1630485.0,
      "step": 725
    },
    {
      "entropy": 6.838800954818725,
      "epoch": 0.7896160086533262,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003645,
      "loss": 6.3953,
      "mean_token_accuracy": 0.13309185206890106,
      "num_tokens": 1641411.0,
      "step": 730
    },
    {
      "entropy": 6.606268787384034,
      "epoch": 0.7950243374797188,
      "grad_norm": 0.953125,
      "learning_rate": 0.000367,
      "loss": 6.4154,
      "mean_token_accuracy": 0.13472312912344933,
      "num_tokens": 1653737.0,
      "step": 735
    },
    {
      "entropy": 6.835287189483642,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003695,
      "loss": 6.5903,
      "mean_token_accuracy": 0.12430674731731414,
      "num_tokens": 1665910.0,
      "step": 740
    },
    {
      "entropy": 6.817589139938354,
      "epoch": 0.8058409951325041,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000372,
      "loss": 6.3509,
      "mean_token_accuracy": 0.1289426304399967,
      "num_tokens": 1676927.0,
      "step": 745
    },
    {
      "entropy": 6.585090637207031,
      "epoch": 0.8112493239588967,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003745,
      "loss": 6.4032,
      "mean_token_accuracy": 0.13011449351906776,
      "num_tokens": 1688167.0,
      "step": 750
    },
    {
      "entropy": 6.720578002929687,
      "epoch": 0.8166576527852893,
      "grad_norm": 0.96875,
      "learning_rate": 0.000377,
      "loss": 6.4835,
      "mean_token_accuracy": 0.13275314942002298,
      "num_tokens": 1699899.0,
      "step": 755
    },
    {
      "entropy": 6.661290502548217,
      "epoch": 0.822065981611682,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003795,
      "loss": 6.4142,
      "mean_token_accuracy": 0.13195080161094666,
      "num_tokens": 1709862.0,
      "step": 760
    },
    {
      "entropy": 6.705444717407227,
      "epoch": 0.8274743104380746,
      "grad_norm": 1.15625,
      "learning_rate": 0.000382,
      "loss": 6.3877,
      "mean_token_accuracy": 0.13449530974030494,
      "num_tokens": 1719902.0,
      "step": 765
    },
    {
      "entropy": 6.709910154342651,
      "epoch": 0.8328826392644673,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003845,
      "loss": 6.4083,
      "mean_token_accuracy": 0.13319506570696832,
      "num_tokens": 1730351.0,
      "step": 770
    },
    {
      "entropy": 6.61332688331604,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3675,
      "mean_token_accuracy": 0.13003436774015426,
      "num_tokens": 1741783.0,
      "step": 775
    },
    {
      "entropy": 6.637337017059326,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.0,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4219,
      "mean_token_accuracy": 0.13431969806551933,
      "num_tokens": 1752952.0,
      "step": 780
    },
    {
      "entropy": 6.77474799156189,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4312,
      "mean_token_accuracy": 0.1275618128478527,
      "num_tokens": 1764014.0,
      "step": 785
    },
    {
      "entropy": 6.689550447463989,
      "epoch": 0.8545159545700378,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3963,
      "mean_token_accuracy": 0.13133953213691713,
      "num_tokens": 1776154.0,
      "step": 790
    },
    {
      "entropy": 6.676612186431885,
      "epoch": 0.8599242833964305,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4523,
      "mean_token_accuracy": 0.1290397100150585,
      "num_tokens": 1787292.0,
      "step": 795
    },
    {
      "entropy": 6.641109752655029,
      "epoch": 0.8653326122228231,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003995,
      "loss": 6.4167,
      "mean_token_accuracy": 0.13453051149845124,
      "num_tokens": 1799469.0,
      "step": 800
    },
    {
      "entropy": 6.551521062850952,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.078125,
      "learning_rate": 0.000402,
      "loss": 6.2944,
      "mean_token_accuracy": 0.13961234614253043,
      "num_tokens": 1810091.0,
      "step": 805
    },
    {
      "entropy": 6.746007585525513,
      "epoch": 0.8761492698756085,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004045,
      "loss": 6.4587,
      "mean_token_accuracy": 0.13312253430485727,
      "num_tokens": 1820835.0,
      "step": 810
    },
    {
      "entropy": 6.6614655494689945,
      "epoch": 0.8815575987020011,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4069,
      "mean_token_accuracy": 0.13295662850141526,
      "num_tokens": 1831868.0,
      "step": 815
    },
    {
      "entropy": 6.575878047943116,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004095,
      "loss": 6.387,
      "mean_token_accuracy": 0.13565733730793,
      "num_tokens": 1842040.0,
      "step": 820
    },
    {
      "entropy": 6.7888940334320065,
      "epoch": 0.8923742563547864,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000412,
      "loss": 6.4603,
      "mean_token_accuracy": 0.13313809186220169,
      "num_tokens": 1853765.0,
      "step": 825
    },
    {
      "entropy": 6.690069341659546,
      "epoch": 0.897782585181179,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004145,
      "loss": 6.4548,
      "mean_token_accuracy": 0.13598668873310088,
      "num_tokens": 1866081.0,
      "step": 830
    },
    {
      "entropy": 6.610725545883179,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000417,
      "loss": 6.364,
      "mean_token_accuracy": 0.1348128743469715,
      "num_tokens": 1876754.0,
      "step": 835
    },
    {
      "entropy": 6.72492733001709,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004195,
      "loss": 6.4677,
      "mean_token_accuracy": 0.12978390827775002,
      "num_tokens": 1888299.0,
      "step": 840
    },
    {
      "entropy": 6.47598352432251,
      "epoch": 0.914007571660357,
      "grad_norm": 1.078125,
      "learning_rate": 0.000422,
      "loss": 6.2533,
      "mean_token_accuracy": 0.1417229875922203,
      "num_tokens": 1899027.0,
      "step": 845
    },
    {
      "entropy": 6.667154026031494,
      "epoch": 0.9194159004867496,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004245,
      "loss": 6.3774,
      "mean_token_accuracy": 0.1327328972518444,
      "num_tokens": 1911043.0,
      "step": 850
    },
    {
      "entropy": 6.636425971984863,
      "epoch": 0.9248242293131422,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000427,
      "loss": 6.3034,
      "mean_token_accuracy": 0.13325904086232185,
      "num_tokens": 1921295.0,
      "step": 855
    },
    {
      "entropy": 6.645001459121704,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004295,
      "loss": 6.4006,
      "mean_token_accuracy": 0.13651155978441237,
      "num_tokens": 1932944.0,
      "step": 860
    },
    {
      "entropy": 6.538773679733277,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432,
      "loss": 6.3351,
      "mean_token_accuracy": 0.13037271574139594,
      "num_tokens": 1944072.0,
      "step": 865
    },
    {
      "entropy": 6.591965293884277,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004345,
      "loss": 6.3428,
      "mean_token_accuracy": 0.13658150658011436,
      "num_tokens": 1954798.0,
      "step": 870
    },
    {
      "entropy": 6.506063842773438,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000437,
      "loss": 6.3039,
      "mean_token_accuracy": 0.1442844972014427,
      "num_tokens": 1965304.0,
      "step": 875
    },
    {
      "entropy": 6.660945653915405,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395,
      "loss": 6.3527,
      "mean_token_accuracy": 0.13428603932261468,
      "num_tokens": 1976797.0,
      "step": 880
    },
    {
      "entropy": 6.56065354347229,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.0,
      "learning_rate": 0.000442,
      "loss": 6.374,
      "mean_token_accuracy": 0.13659664168953894,
      "num_tokens": 1987448.0,
      "step": 885
    },
    {
      "entropy": 6.366022777557373,
      "epoch": 0.9626825310978907,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004445,
      "loss": 6.2006,
      "mean_token_accuracy": 0.13602133318781853,
      "num_tokens": 2000948.0,
      "step": 890
    },
    {
      "entropy": 6.597631502151489,
      "epoch": 0.9680908599242833,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000447,
      "loss": 6.3109,
      "mean_token_accuracy": 0.13974019661545753,
      "num_tokens": 2011807.0,
      "step": 895
    },
    {
      "entropy": 6.633898687362671,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3585,
      "mean_token_accuracy": 0.13105632588267327,
      "num_tokens": 2022835.0,
      "step": 900
    },
    {
      "entropy": 6.387359571456909,
      "epoch": 0.9789075175770687,
      "grad_norm": 0.875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.276,
      "mean_token_accuracy": 0.14153993576765062,
      "num_tokens": 2034331.0,
      "step": 905
    },
    {
      "entropy": 6.695045137405396,
      "epoch": 0.9843158464034614,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2744,
      "mean_token_accuracy": 0.13374185785651208,
      "num_tokens": 2047002.0,
      "step": 910
    },
    {
      "entropy": 6.452011156082153,
      "epoch": 0.989724175229854,
      "grad_norm": 1.0,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2301,
      "mean_token_accuracy": 0.14060857594013215,
      "num_tokens": 2056299.0,
      "step": 915
    },
    {
      "entropy": 6.562105369567871,
      "epoch": 0.9951325040562466,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3512,
      "mean_token_accuracy": 0.13779174089431762,
      "num_tokens": 2068031.0,
      "step": 920
    },
    {
      "entropy": 6.541797320048015,
      "epoch": 1.0,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000462,
      "loss": 6.2951,
      "mean_token_accuracy": 0.14415054851108128,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.501549148559571,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.1676,
      "mean_token_accuracy": 0.14561108872294426,
      "num_tokens": 2087316.0,
      "step": 930
    },
    {
      "entropy": 6.5583666324615475,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.015625,
      "learning_rate": 0.000467,
      "loss": 6.1739,
      "mean_token_accuracy": 0.13918173015117646,
      "num_tokens": 2098522.0,
      "step": 935
    },
    {
      "entropy": 6.392220306396484,
      "epoch": 1.016224986479178,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004695,
      "loss": 6.1354,
      "mean_token_accuracy": 0.14603151232004166,
      "num_tokens": 2110015.0,
      "step": 940
    },
    {
      "entropy": 6.509898853302002,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.015625,
      "learning_rate": 0.000472,
      "loss": 6.2538,
      "mean_token_accuracy": 0.13781652599573135,
      "num_tokens": 2121333.0,
      "step": 945
    },
    {
      "entropy": 6.577088499069214,
      "epoch": 1.0270416441319632,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004745,
      "loss": 6.138,
      "mean_token_accuracy": 0.1490761384367943,
      "num_tokens": 2131385.0,
      "step": 950
    },
    {
      "entropy": 6.454202699661255,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477,
      "loss": 6.1985,
      "mean_token_accuracy": 0.1469516322016716,
      "num_tokens": 2143274.0,
      "step": 955
    },
    {
      "entropy": 6.380458831787109,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004795,
      "loss": 6.0657,
      "mean_token_accuracy": 0.14620195776224137,
      "num_tokens": 2154529.0,
      "step": 960
    },
    {
      "entropy": 6.4978209972381595,
      "epoch": 1.043266630611141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000482,
      "loss": 6.2219,
      "mean_token_accuracy": 0.14443108439445496,
      "num_tokens": 2165949.0,
      "step": 965
    },
    {
      "entropy": 6.401442909240723,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004845,
      "loss": 6.1906,
      "mean_token_accuracy": 0.14745446890592576,
      "num_tokens": 2176710.0,
      "step": 970
    },
    {
      "entropy": 6.489724588394165,
      "epoch": 1.0540832882639264,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487,
      "loss": 6.1687,
      "mean_token_accuracy": 0.15010684430599214,
      "num_tokens": 2189470.0,
      "step": 975
    },
    {
      "entropy": 6.376638793945313,
      "epoch": 1.059491617090319,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895,
      "loss": 6.1227,
      "mean_token_accuracy": 0.15111407190561293,
      "num_tokens": 2200266.0,
      "step": 980
    },
    {
      "entropy": 6.467359828948974,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492,
      "loss": 6.1322,
      "mean_token_accuracy": 0.1447245694696903,
      "num_tokens": 2210736.0,
      "step": 985
    },
    {
      "entropy": 6.49309549331665,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945,
      "loss": 6.2527,
      "mean_token_accuracy": 0.14038581252098084,
      "num_tokens": 2221057.0,
      "step": 990
    },
    {
      "entropy": 6.4171429634094235,
      "epoch": 1.075716603569497,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497,
      "loss": 6.2282,
      "mean_token_accuracy": 0.14197226315736772,
      "num_tokens": 2232352.0,
      "step": 995
    },
    {
      "entropy": 6.583332157135009,
      "epoch": 1.0811249323958896,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995,
      "loss": 6.134,
      "mean_token_accuracy": 0.149154694378376,
      "num_tokens": 2244655.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 6.15572995597666,
      "eval_loss": 6.430820941925049,
      "eval_mean_token_accuracy": 0.14240490521558308,
      "eval_num_tokens": 2244655.0,
      "eval_runtime": 1.9399,
      "eval_samples_per_second": 1811.932,
      "eval_steps_per_second": 226.814,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 517821876633600.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}