{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9789525208027411,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691474437713623,
      "epoch": 0.004894762604013705,
      "grad_norm": 11.875,
      "learning_rate": 2e-06,
      "loss": 10.8265,
      "mean_token_accuracy": 7.836990407668054e-05,
      "num_tokens": 12593.0,
      "step": 5
    },
    {
      "entropy": 10.691561222076416,
      "epoch": 0.00978952520802741,
      "grad_norm": 11.75,
      "learning_rate": 4.5e-06,
      "loss": 10.7489,
      "mean_token_accuracy": 8.1300811143592e-05,
      "num_tokens": 25410.0,
      "step": 10
    },
    {
      "entropy": 10.691450119018555,
      "epoch": 0.014684287812041116,
      "grad_norm": 9.6875,
      "learning_rate": 7e-06,
      "loss": 10.5176,
      "mean_token_accuracy": 0.009127140190685168,
      "num_tokens": 37264.0,
      "step": 15
    },
    {
      "entropy": 10.688244438171386,
      "epoch": 0.01957905041605482,
      "grad_norm": 6.15625,
      "learning_rate": 9.5e-06,
      "loss": 10.2549,
      "mean_token_accuracy": 0.030842235311865807,
      "num_tokens": 49483.0,
      "step": 20
    },
    {
      "entropy": 10.67002468109131,
      "epoch": 0.024473813020068527,
      "grad_norm": 4.15625,
      "learning_rate": 1.2e-05,
      "loss": 9.9838,
      "mean_token_accuracy": 0.036930659972131255,
      "num_tokens": 61479.0,
      "step": 25
    },
    {
      "entropy": 10.649608993530274,
      "epoch": 0.02936857562408223,
      "grad_norm": 3.28125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8397,
      "mean_token_accuracy": 0.036012667790055276,
      "num_tokens": 73344.0,
      "step": 30
    },
    {
      "entropy": 10.640453147888184,
      "epoch": 0.034263338228095935,
      "grad_norm": 2.734375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.7965,
      "mean_token_accuracy": 0.04287212081253529,
      "num_tokens": 85878.0,
      "step": 35
    },
    {
      "entropy": 10.626565551757812,
      "epoch": 0.03915810083210964,
      "grad_norm": 2.4375,
      "learning_rate": 1.95e-05,
      "loss": 9.7691,
      "mean_token_accuracy": 0.04283439628779888,
      "num_tokens": 97700.0,
      "step": 40
    },
    {
      "entropy": 10.62145414352417,
      "epoch": 0.04405286343612335,
      "grad_norm": 2.421875,
      "learning_rate": 2.2e-05,
      "loss": 9.6755,
      "mean_token_accuracy": 0.04480269178748131,
      "num_tokens": 109008.0,
      "step": 45
    },
    {
      "entropy": 10.633209228515625,
      "epoch": 0.048947626040137054,
      "grad_norm": 2.40625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6612,
      "mean_token_accuracy": 0.04066384620964527,
      "num_tokens": 120635.0,
      "step": 50
    },
    {
      "entropy": 10.635868644714355,
      "epoch": 0.05384238864415076,
      "grad_norm": 2.421875,
      "learning_rate": 2.7e-05,
      "loss": 9.5963,
      "mean_token_accuracy": 0.04003451876342297,
      "num_tokens": 133102.0,
      "step": 55
    },
    {
      "entropy": 10.61568660736084,
      "epoch": 0.05873715124816446,
      "grad_norm": 2.5,
      "learning_rate": 2.95e-05,
      "loss": 9.4728,
      "mean_token_accuracy": 0.050173386186361316,
      "num_tokens": 145445.0,
      "step": 60
    },
    {
      "entropy": 10.51130256652832,
      "epoch": 0.06363191385217817,
      "grad_norm": 2.6875,
      "learning_rate": 3.2e-05,
      "loss": 9.3842,
      "mean_token_accuracy": 0.05366942435503006,
      "num_tokens": 157594.0,
      "step": 65
    },
    {
      "entropy": 10.435814762115479,
      "epoch": 0.06852667645619187,
      "grad_norm": 2.390625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2864,
      "mean_token_accuracy": 0.06021526530385017,
      "num_tokens": 170976.0,
      "step": 70
    },
    {
      "entropy": 10.467036819458007,
      "epoch": 0.07342143906020558,
      "grad_norm": 2.875,
      "learning_rate": 3.7e-05,
      "loss": 9.2122,
      "mean_token_accuracy": 0.06290490776300431,
      "num_tokens": 183561.0,
      "step": 75
    },
    {
      "entropy": 10.445936489105225,
      "epoch": 0.07831620166421928,
      "grad_norm": 2.265625,
      "learning_rate": 3.95e-05,
      "loss": 9.0845,
      "mean_token_accuracy": 0.07000090889632701,
      "num_tokens": 195936.0,
      "step": 80
    },
    {
      "entropy": 10.329971313476562,
      "epoch": 0.08321096426823299,
      "grad_norm": 2.5,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.9658,
      "mean_token_accuracy": 0.07175729237496853,
      "num_tokens": 208504.0,
      "step": 85
    },
    {
      "entropy": 10.371177291870117,
      "epoch": 0.0881057268722467,
      "grad_norm": 2.484375,
      "learning_rate": 4.45e-05,
      "loss": 8.9385,
      "mean_token_accuracy": 0.06789587959647178,
      "num_tokens": 219722.0,
      "step": 90
    },
    {
      "entropy": 10.313321971893311,
      "epoch": 0.0930004894762604,
      "grad_norm": 2.21875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.7436,
      "mean_token_accuracy": 0.07605377212166786,
      "num_tokens": 232307.0,
      "step": 95
    },
    {
      "entropy": 10.276076507568359,
      "epoch": 0.09789525208027411,
      "grad_norm": 3.28125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.6775,
      "mean_token_accuracy": 0.07455896399915218,
      "num_tokens": 243698.0,
      "step": 100
    },
    {
      "entropy": 10.196091747283935,
      "epoch": 0.1027900146842878,
      "grad_norm": 2.390625,
      "learning_rate": 5.2e-05,
      "loss": 8.5575,
      "mean_token_accuracy": 0.0785155214369297,
      "num_tokens": 256418.0,
      "step": 105
    },
    {
      "entropy": 10.124401664733886,
      "epoch": 0.10768477728830152,
      "grad_norm": 2.40625,
      "learning_rate": 5.45e-05,
      "loss": 8.4104,
      "mean_token_accuracy": 0.07805168032646179,
      "num_tokens": 268503.0,
      "step": 110
    },
    {
      "entropy": 10.06371488571167,
      "epoch": 0.11257953989231523,
      "grad_norm": 2.21875,
      "learning_rate": 5.7e-05,
      "loss": 8.3101,
      "mean_token_accuracy": 0.07913323007524013,
      "num_tokens": 280386.0,
      "step": 115
    },
    {
      "entropy": 9.980810928344727,
      "epoch": 0.11747430249632893,
      "grad_norm": 2.296875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2019,
      "mean_token_accuracy": 0.08205665312707425,
      "num_tokens": 292121.0,
      "step": 120
    },
    {
      "entropy": 9.89662504196167,
      "epoch": 0.12236906510034264,
      "grad_norm": 2.1875,
      "learning_rate": 6.2e-05,
      "loss": 8.0286,
      "mean_token_accuracy": 0.08492446094751357,
      "num_tokens": 303782.0,
      "step": 125
    },
    {
      "entropy": 9.724525547027588,
      "epoch": 0.12726382770435635,
      "grad_norm": 1.8671875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9844,
      "mean_token_accuracy": 0.08128817304968834,
      "num_tokens": 315497.0,
      "step": 130
    },
    {
      "entropy": 9.641802310943604,
      "epoch": 0.13215859030837004,
      "grad_norm": 1.6875,
      "learning_rate": 6.7e-05,
      "loss": 7.9532,
      "mean_token_accuracy": 0.07768743187189102,
      "num_tokens": 327840.0,
      "step": 135
    },
    {
      "entropy": 9.46103801727295,
      "epoch": 0.13705335291238374,
      "grad_norm": 1.7421875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7654,
      "mean_token_accuracy": 0.08798187449574471,
      "num_tokens": 340192.0,
      "step": 140
    },
    {
      "entropy": 9.228885173797607,
      "epoch": 0.14194811551639747,
      "grad_norm": 1.9140625,
      "learning_rate": 7.2e-05,
      "loss": 7.5669,
      "mean_token_accuracy": 0.08950990438461304,
      "num_tokens": 351561.0,
      "step": 145
    },
    {
      "entropy": 9.089824771881103,
      "epoch": 0.14684287812041116,
      "grad_norm": 1.6328125,
      "learning_rate": 7.45e-05,
      "loss": 7.5159,
      "mean_token_accuracy": 0.08800971060991288,
      "num_tokens": 363668.0,
      "step": 150
    },
    {
      "entropy": 8.857299709320069,
      "epoch": 0.15173764072442486,
      "grad_norm": 1.6953125,
      "learning_rate": 7.7e-05,
      "loss": 7.4525,
      "mean_token_accuracy": 0.09529466852545738,
      "num_tokens": 375241.0,
      "step": 155
    },
    {
      "entropy": 8.673880004882813,
      "epoch": 0.15663240332843856,
      "grad_norm": 1.90625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.3172,
      "mean_token_accuracy": 0.09652032777667045,
      "num_tokens": 386488.0,
      "step": 160
    },
    {
      "entropy": 8.434674930572509,
      "epoch": 0.16152716593245228,
      "grad_norm": 1.5078125,
      "learning_rate": 8.2e-05,
      "loss": 7.2006,
      "mean_token_accuracy": 0.09962413907051086,
      "num_tokens": 398203.0,
      "step": 165
    },
    {
      "entropy": 8.35833911895752,
      "epoch": 0.16642192853646598,
      "grad_norm": 1.59375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.2842,
      "mean_token_accuracy": 0.09556651711463929,
      "num_tokens": 409381.0,
      "step": 170
    },
    {
      "entropy": 8.216682815551758,
      "epoch": 0.17131669114047968,
      "grad_norm": 1.5625,
      "learning_rate": 8.7e-05,
      "loss": 7.2633,
      "mean_token_accuracy": 0.09789874404668808,
      "num_tokens": 421760.0,
      "step": 175
    },
    {
      "entropy": 8.080347347259522,
      "epoch": 0.1762114537444934,
      "grad_norm": 1.46875,
      "learning_rate": 8.95e-05,
      "loss": 7.1632,
      "mean_token_accuracy": 0.10209440737962723,
      "num_tokens": 432727.0,
      "step": 180
    },
    {
      "entropy": 8.032786130905151,
      "epoch": 0.1811062163485071,
      "grad_norm": 1.84375,
      "learning_rate": 9.2e-05,
      "loss": 7.2471,
      "mean_token_accuracy": 0.09438547641038894,
      "num_tokens": 443799.0,
      "step": 185
    },
    {
      "entropy": 7.938335800170899,
      "epoch": 0.1860009789525208,
      "grad_norm": 1.3359375,
      "learning_rate": 9.45e-05,
      "loss": 7.1341,
      "mean_token_accuracy": 0.10013610497117043,
      "num_tokens": 455481.0,
      "step": 190
    },
    {
      "entropy": 7.9153131484985355,
      "epoch": 0.19089574155653452,
      "grad_norm": 1.625,
      "learning_rate": 9.7e-05,
      "loss": 7.0763,
      "mean_token_accuracy": 0.09839447140693665,
      "num_tokens": 467086.0,
      "step": 195
    },
    {
      "entropy": 7.842766427993775,
      "epoch": 0.19579050416054822,
      "grad_norm": 1.640625,
      "learning_rate": 9.95e-05,
      "loss": 7.0763,
      "mean_token_accuracy": 0.09429099857807159,
      "num_tokens": 478227.0,
      "step": 200
    },
    {
      "entropy": 7.8508209705352785,
      "epoch": 0.20068526676456191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000102,
      "loss": 7.0819,
      "mean_token_accuracy": 0.10272094458341599,
      "num_tokens": 491036.0,
      "step": 205
    },
    {
      "entropy": 7.750106430053711,
      "epoch": 0.2055800293685756,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0268,
      "mean_token_accuracy": 0.10150075927376748,
      "num_tokens": 502833.0,
      "step": 210
    },
    {
      "entropy": 7.662298250198364,
      "epoch": 0.21047479197258934,
      "grad_norm": 1.53125,
      "learning_rate": 0.000107,
      "loss": 6.9994,
      "mean_token_accuracy": 0.09913612380623818,
      "num_tokens": 514078.0,
      "step": 215
    },
    {
      "entropy": 7.76820936203003,
      "epoch": 0.21536955457660303,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001095,
      "loss": 7.0563,
      "mean_token_accuracy": 0.09886371418833732,
      "num_tokens": 525505.0,
      "step": 220
    },
    {
      "entropy": 7.7181964874267575,
      "epoch": 0.22026431718061673,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000112,
      "loss": 7.0553,
      "mean_token_accuracy": 0.10015972703695297,
      "num_tokens": 536684.0,
      "step": 225
    },
    {
      "entropy": 7.631637525558472,
      "epoch": 0.22515907978463046,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0001145,
      "loss": 6.9678,
      "mean_token_accuracy": 0.11216415911912918,
      "num_tokens": 547664.0,
      "step": 230
    },
    {
      "entropy": 7.560673618316651,
      "epoch": 0.23005384238864415,
      "grad_norm": 1.78125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9656,
      "mean_token_accuracy": 0.10480529963970184,
      "num_tokens": 559892.0,
      "step": 235
    },
    {
      "entropy": 7.457711982727051,
      "epoch": 0.23494860499265785,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.9001,
      "mean_token_accuracy": 0.10354893431067466,
      "num_tokens": 571012.0,
      "step": 240
    },
    {
      "entropy": 7.583689403533936,
      "epoch": 0.23984336759667158,
      "grad_norm": 1.609375,
      "learning_rate": 0.000122,
      "loss": 6.9215,
      "mean_token_accuracy": 0.10430897325277329,
      "num_tokens": 583451.0,
      "step": 245
    },
    {
      "entropy": 7.421637296676636,
      "epoch": 0.24473813020068527,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001245,
      "loss": 6.9329,
      "mean_token_accuracy": 0.10233990103006363,
      "num_tokens": 595915.0,
      "step": 250
    },
    {
      "entropy": 7.562687349319458,
      "epoch": 0.24963289280469897,
      "grad_norm": 1.359375,
      "learning_rate": 0.000127,
      "loss": 6.848,
      "mean_token_accuracy": 0.10910093262791634,
      "num_tokens": 607863.0,
      "step": 255
    },
    {
      "entropy": 7.433994293212891,
      "epoch": 0.2545276554087127,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001295,
      "loss": 6.9084,
      "mean_token_accuracy": 0.10604727566242218,
      "num_tokens": 619306.0,
      "step": 260
    },
    {
      "entropy": 7.461258888244629,
      "epoch": 0.25942241801272636,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000132,
      "loss": 6.9468,
      "mean_token_accuracy": 0.10922192186117172,
      "num_tokens": 632553.0,
      "step": 265
    },
    {
      "entropy": 7.41155481338501,
      "epoch": 0.2643171806167401,
      "grad_norm": 1.671875,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.7688,
      "mean_token_accuracy": 0.11494352146983147,
      "num_tokens": 643399.0,
      "step": 270
    },
    {
      "entropy": 7.398804092407227,
      "epoch": 0.2692119432207538,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8403,
      "mean_token_accuracy": 0.1158620223402977,
      "num_tokens": 654148.0,
      "step": 275
    },
    {
      "entropy": 7.385111284255982,
      "epoch": 0.2741067058247675,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001395,
      "loss": 6.8965,
      "mean_token_accuracy": 0.11017908081412316,
      "num_tokens": 666324.0,
      "step": 280
    },
    {
      "entropy": 7.312957096099853,
      "epoch": 0.2790014684287812,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8106,
      "mean_token_accuracy": 0.11318352445960045,
      "num_tokens": 677309.0,
      "step": 285
    },
    {
      "entropy": 7.313508749008179,
      "epoch": 0.28389623103279493,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001445,
      "loss": 6.8144,
      "mean_token_accuracy": 0.10523941665887833,
      "num_tokens": 691469.0,
      "step": 290
    },
    {
      "entropy": 7.350642156600952,
      "epoch": 0.2887909936368086,
      "grad_norm": 1.546875,
      "learning_rate": 0.000147,
      "loss": 6.8045,
      "mean_token_accuracy": 0.12274143844842911,
      "num_tokens": 702985.0,
      "step": 295
    },
    {
      "entropy": 7.323756074905395,
      "epoch": 0.2936857562408223,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001495,
      "loss": 6.8789,
      "mean_token_accuracy": 0.116054268181324,
      "num_tokens": 715484.0,
      "step": 300
    },
    {
      "entropy": 7.248594284057617,
      "epoch": 0.29858051884483605,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000152,
      "loss": 6.7608,
      "mean_token_accuracy": 0.11559304669499397,
      "num_tokens": 725967.0,
      "step": 305
    },
    {
      "entropy": 7.158112096786499,
      "epoch": 0.3034752814488497,
      "grad_norm": 1.671875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.7019,
      "mean_token_accuracy": 0.12221056818962098,
      "num_tokens": 737878.0,
      "step": 310
    },
    {
      "entropy": 7.245487976074219,
      "epoch": 0.30837004405286345,
      "grad_norm": 1.375,
      "learning_rate": 0.000157,
      "loss": 6.7697,
      "mean_token_accuracy": 0.1148179180920124,
      "num_tokens": 750444.0,
      "step": 315
    },
    {
      "entropy": 7.131648015975952,
      "epoch": 0.3132648066568771,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001595,
      "loss": 6.7355,
      "mean_token_accuracy": 0.11609084904193878,
      "num_tokens": 761905.0,
      "step": 320
    },
    {
      "entropy": 7.189060926437378,
      "epoch": 0.31815956926089084,
      "grad_norm": 1.25,
      "learning_rate": 0.000162,
      "loss": 6.6511,
      "mean_token_accuracy": 0.11201457530260087,
      "num_tokens": 773238.0,
      "step": 325
    },
    {
      "entropy": 7.255410861968994,
      "epoch": 0.32305433186490456,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.818,
      "mean_token_accuracy": 0.11478052586317063,
      "num_tokens": 785265.0,
      "step": 330
    },
    {
      "entropy": 7.000382041931152,
      "epoch": 0.32794909446891823,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.5323,
      "mean_token_accuracy": 0.12431756779551506,
      "num_tokens": 796214.0,
      "step": 335
    },
    {
      "entropy": 7.132069635391235,
      "epoch": 0.33284385707293196,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7048,
      "mean_token_accuracy": 0.12211181521415711,
      "num_tokens": 807429.0,
      "step": 340
    },
    {
      "entropy": 7.093725252151489,
      "epoch": 0.3377386196769457,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7147,
      "mean_token_accuracy": 0.11440666690468788,
      "num_tokens": 819205.0,
      "step": 345
    },
    {
      "entropy": 7.087794494628906,
      "epoch": 0.34263338228095935,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.6921,
      "mean_token_accuracy": 0.11648698076605797,
      "num_tokens": 831417.0,
      "step": 350
    },
    {
      "entropy": 7.148236560821533,
      "epoch": 0.3475281448849731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000177,
      "loss": 6.7295,
      "mean_token_accuracy": 0.11438237279653549,
      "num_tokens": 844792.0,
      "step": 355
    },
    {
      "entropy": 7.161906480789185,
      "epoch": 0.3524229074889868,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001795,
      "loss": 6.7167,
      "mean_token_accuracy": 0.1205228291451931,
      "num_tokens": 856643.0,
      "step": 360
    },
    {
      "entropy": 7.08436393737793,
      "epoch": 0.3573176700930005,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000182,
      "loss": 6.6894,
      "mean_token_accuracy": 0.1218317560851574,
      "num_tokens": 867727.0,
      "step": 365
    },
    {
      "entropy": 7.0502026081085205,
      "epoch": 0.3622124326970142,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001845,
      "loss": 6.6484,
      "mean_token_accuracy": 0.12172565683722496,
      "num_tokens": 879580.0,
      "step": 370
    },
    {
      "entropy": 7.093457746505737,
      "epoch": 0.3671071953010279,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000187,
      "loss": 6.6455,
      "mean_token_accuracy": 0.11903778240084648,
      "num_tokens": 890603.0,
      "step": 375
    },
    {
      "entropy": 7.022216749191284,
      "epoch": 0.3720019579050416,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001895,
      "loss": 6.6953,
      "mean_token_accuracy": 0.11619957983493805,
      "num_tokens": 902566.0,
      "step": 380
    },
    {
      "entropy": 7.156719017028808,
      "epoch": 0.3768967205090553,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000192,
      "loss": 6.7699,
      "mean_token_accuracy": 0.11908866539597511,
      "num_tokens": 915158.0,
      "step": 385
    },
    {
      "entropy": 7.020287132263183,
      "epoch": 0.38179148311306904,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001945,
      "loss": 6.6569,
      "mean_token_accuracy": 0.12016270384192466,
      "num_tokens": 928299.0,
      "step": 390
    },
    {
      "entropy": 6.987773513793945,
      "epoch": 0.3866862457170827,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7094,
      "mean_token_accuracy": 0.11790777072310447,
      "num_tokens": 940470.0,
      "step": 395
    },
    {
      "entropy": 7.102205657958985,
      "epoch": 0.39158100832109644,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6851,
      "mean_token_accuracy": 0.12131372764706612,
      "num_tokens": 952661.0,
      "step": 400
    },
    {
      "entropy": 6.911374616622925,
      "epoch": 0.3964757709251101,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000202,
      "loss": 6.6155,
      "mean_token_accuracy": 0.11879359632730484,
      "num_tokens": 965040.0,
      "step": 405
    },
    {
      "entropy": 6.855518102645874,
      "epoch": 0.40137053352912383,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.5407,
      "mean_token_accuracy": 0.12074877321720123,
      "num_tokens": 977655.0,
      "step": 410
    },
    {
      "entropy": 7.038438987731934,
      "epoch": 0.40626529613313755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000207,
      "loss": 6.6291,
      "mean_token_accuracy": 0.12360285148024559,
      "num_tokens": 989422.0,
      "step": 415
    },
    {
      "entropy": 6.930527019500732,
      "epoch": 0.4111600587371512,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002095,
      "loss": 6.6136,
      "mean_token_accuracy": 0.1211114652454853,
      "num_tokens": 1002697.0,
      "step": 420
    },
    {
      "entropy": 6.963910531997681,
      "epoch": 0.41605482134116495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000212,
      "loss": 6.5715,
      "mean_token_accuracy": 0.12396370768547058,
      "num_tokens": 1014910.0,
      "step": 425
    },
    {
      "entropy": 6.838316869735718,
      "epoch": 0.4209495839451787,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002145,
      "loss": 6.5537,
      "mean_token_accuracy": 0.1339319072663784,
      "num_tokens": 1026835.0,
      "step": 430
    },
    {
      "entropy": 7.047098207473755,
      "epoch": 0.42584434654919234,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6707,
      "mean_token_accuracy": 0.11443433165550232,
      "num_tokens": 1038792.0,
      "step": 435
    },
    {
      "entropy": 6.900348567962647,
      "epoch": 0.43073910915320607,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002195,
      "loss": 6.6413,
      "mean_token_accuracy": 0.11995069831609725,
      "num_tokens": 1051335.0,
      "step": 440
    },
    {
      "entropy": 6.856663560867309,
      "epoch": 0.4356338717572198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000222,
      "loss": 6.5079,
      "mean_token_accuracy": 0.130057705193758,
      "num_tokens": 1063480.0,
      "step": 445
    },
    {
      "entropy": 6.915047836303711,
      "epoch": 0.44052863436123346,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002245,
      "loss": 6.6642,
      "mean_token_accuracy": 0.12786006629467012,
      "num_tokens": 1074925.0,
      "step": 450
    },
    {
      "entropy": 6.8480541706085205,
      "epoch": 0.4454233969652472,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.396,
      "mean_token_accuracy": 0.12921193838119507,
      "num_tokens": 1086068.0,
      "step": 455
    },
    {
      "entropy": 6.727921867370606,
      "epoch": 0.4503181595692609,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.4969,
      "mean_token_accuracy": 0.13089798390865326,
      "num_tokens": 1097309.0,
      "step": 460
    },
    {
      "entropy": 6.8872028350830075,
      "epoch": 0.4552129221732746,
      "grad_norm": 1.375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5581,
      "mean_token_accuracy": 0.125684704631567,
      "num_tokens": 1110253.0,
      "step": 465
    },
    {
      "entropy": 6.802498722076416,
      "epoch": 0.4601076847772883,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5386,
      "mean_token_accuracy": 0.125102449208498,
      "num_tokens": 1121297.0,
      "step": 470
    },
    {
      "entropy": 6.830904102325439,
      "epoch": 0.46500244738130203,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000237,
      "loss": 6.4694,
      "mean_token_accuracy": 0.13397082462906837,
      "num_tokens": 1133378.0,
      "step": 475
    },
    {
      "entropy": 6.764972782135009,
      "epoch": 0.4698972099853157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002395,
      "loss": 6.4908,
      "mean_token_accuracy": 0.1247988536953926,
      "num_tokens": 1146197.0,
      "step": 480
    },
    {
      "entropy": 6.834934377670288,
      "epoch": 0.4747919725893294,
      "grad_norm": 1.484375,
      "learning_rate": 0.000242,
      "loss": 6.4895,
      "mean_token_accuracy": 0.12815890833735466,
      "num_tokens": 1158791.0,
      "step": 485
    },
    {
      "entropy": 6.834185266494751,
      "epoch": 0.47968673519334315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002445,
      "loss": 6.4552,
      "mean_token_accuracy": 0.13044697493314744,
      "num_tokens": 1169977.0,
      "step": 490
    },
    {
      "entropy": 6.6346056938171385,
      "epoch": 0.4845814977973568,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247,
      "loss": 6.4207,
      "mean_token_accuracy": 0.13247148767113687,
      "num_tokens": 1182825.0,
      "step": 495
    },
    {
      "entropy": 6.824855661392212,
      "epoch": 0.48947626040137054,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002495,
      "loss": 6.5415,
      "mean_token_accuracy": 0.12181857749819755,
      "num_tokens": 1196101.0,
      "step": 500
    },
    {
      "entropy": 6.662319135665894,
      "epoch": 0.4943710230053842,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000252,
      "loss": 6.3397,
      "mean_token_accuracy": 0.1353870265185833,
      "num_tokens": 1208916.0,
      "step": 505
    },
    {
      "entropy": 6.704164838790893,
      "epoch": 0.49926578560939794,
      "grad_norm": 1.375,
      "learning_rate": 0.0002545,
      "loss": 6.3986,
      "mean_token_accuracy": 0.13314119651913642,
      "num_tokens": 1218972.0,
      "step": 510
    },
    {
      "entropy": 6.779240608215332,
      "epoch": 0.5041605482134116,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000257,
      "loss": 6.4141,
      "mean_token_accuracy": 0.13696231916546822,
      "num_tokens": 1230692.0,
      "step": 515
    },
    {
      "entropy": 6.710984706878662,
      "epoch": 0.5090553108174254,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002595,
      "loss": 6.3371,
      "mean_token_accuracy": 0.14076263681054116,
      "num_tokens": 1242104.0,
      "step": 520
    },
    {
      "entropy": 6.6697978496551515,
      "epoch": 0.5139500734214391,
      "grad_norm": 1.15625,
      "learning_rate": 0.000262,
      "loss": 6.3853,
      "mean_token_accuracy": 0.1359956182539463,
      "num_tokens": 1254119.0,
      "step": 525
    },
    {
      "entropy": 6.737369012832642,
      "epoch": 0.5188448360254527,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.4548,
      "mean_token_accuracy": 0.1344248741865158,
      "num_tokens": 1265272.0,
      "step": 530
    },
    {
      "entropy": 6.730101156234741,
      "epoch": 0.5237395986294665,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.4222,
      "mean_token_accuracy": 0.14020417779684066,
      "num_tokens": 1277564.0,
      "step": 535
    },
    {
      "entropy": 6.691190624237061,
      "epoch": 0.5286343612334802,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.4491,
      "mean_token_accuracy": 0.13473157808184624,
      "num_tokens": 1289021.0,
      "step": 540
    },
    {
      "entropy": 6.646734714508057,
      "epoch": 0.5335291238374938,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.3956,
      "mean_token_accuracy": 0.13235612735152244,
      "num_tokens": 1301101.0,
      "step": 545
    },
    {
      "entropy": 6.719683647155762,
      "epoch": 0.5384238864415076,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002745,
      "loss": 6.5075,
      "mean_token_accuracy": 0.1343328684568405,
      "num_tokens": 1312982.0,
      "step": 550
    },
    {
      "entropy": 6.649339962005615,
      "epoch": 0.5433186490455213,
      "grad_norm": 1.296875,
      "learning_rate": 0.000277,
      "loss": 6.298,
      "mean_token_accuracy": 0.13852434232831,
      "num_tokens": 1323587.0,
      "step": 555
    },
    {
      "entropy": 6.660077142715454,
      "epoch": 0.548213411649535,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002795,
      "loss": 6.4452,
      "mean_token_accuracy": 0.13839047104120256,
      "num_tokens": 1335989.0,
      "step": 560
    },
    {
      "entropy": 6.759215688705444,
      "epoch": 0.5531081742535487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.47,
      "mean_token_accuracy": 0.130010923743248,
      "num_tokens": 1348065.0,
      "step": 565
    },
    {
      "entropy": 6.673825311660766,
      "epoch": 0.5580029368575624,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002845,
      "loss": 6.3943,
      "mean_token_accuracy": 0.13619856983423234,
      "num_tokens": 1360005.0,
      "step": 570
    },
    {
      "entropy": 6.64725513458252,
      "epoch": 0.5628976994615761,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000287,
      "loss": 6.4037,
      "mean_token_accuracy": 0.13103986009955407,
      "num_tokens": 1372044.0,
      "step": 575
    },
    {
      "entropy": 6.6476539134979244,
      "epoch": 0.5677924620655899,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002895,
      "loss": 6.4819,
      "mean_token_accuracy": 0.13560598343610764,
      "num_tokens": 1384078.0,
      "step": 580
    },
    {
      "entropy": 6.72802586555481,
      "epoch": 0.5726872246696035,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000292,
      "loss": 6.3781,
      "mean_token_accuracy": 0.1315067872405052,
      "num_tokens": 1396102.0,
      "step": 585
    },
    {
      "entropy": 6.672314643859863,
      "epoch": 0.5775819872736172,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002945,
      "loss": 6.366,
      "mean_token_accuracy": 0.13516665622591972,
      "num_tokens": 1408009.0,
      "step": 590
    },
    {
      "entropy": 6.647306442260742,
      "epoch": 0.582476749877631,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000297,
      "loss": 6.3929,
      "mean_token_accuracy": 0.13240321800112725,
      "num_tokens": 1419883.0,
      "step": 595
    },
    {
      "entropy": 6.443280220031738,
      "epoch": 0.5873715124816447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002995,
      "loss": 6.1976,
      "mean_token_accuracy": 0.1442122057080269,
      "num_tokens": 1430901.0,
      "step": 600
    },
    {
      "entropy": 6.578019142150879,
      "epoch": 0.5922662750856583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000302,
      "loss": 6.3251,
      "mean_token_accuracy": 0.13777725845575334,
      "num_tokens": 1443614.0,
      "step": 605
    },
    {
      "entropy": 6.604838514328003,
      "epoch": 0.5971610376896721,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003045,
      "loss": 6.3186,
      "mean_token_accuracy": 0.13596051931381226,
      "num_tokens": 1454750.0,
      "step": 610
    },
    {
      "entropy": 6.47489013671875,
      "epoch": 0.6020558002936858,
      "grad_norm": 1.109375,
      "learning_rate": 0.000307,
      "loss": 6.263,
      "mean_token_accuracy": 0.13405164256691932,
      "num_tokens": 1466424.0,
      "step": 615
    },
    {
      "entropy": 6.579163694381714,
      "epoch": 0.6069505628976994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003095,
      "loss": 6.3106,
      "mean_token_accuracy": 0.1360399268567562,
      "num_tokens": 1478762.0,
      "step": 620
    },
    {
      "entropy": 6.509273624420166,
      "epoch": 0.6118453255017132,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000312,
      "loss": 6.3347,
      "mean_token_accuracy": 0.13646816983819007,
      "num_tokens": 1489683.0,
      "step": 625
    },
    {
      "entropy": 6.514836597442627,
      "epoch": 0.6167400881057269,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003145,
      "loss": 6.2377,
      "mean_token_accuracy": 0.13515424355864525,
      "num_tokens": 1503307.0,
      "step": 630
    },
    {
      "entropy": 6.6227192878723145,
      "epoch": 0.6216348507097406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000317,
      "loss": 6.3727,
      "mean_token_accuracy": 0.13602764680981635,
      "num_tokens": 1514982.0,
      "step": 635
    },
    {
      "entropy": 6.397299718856812,
      "epoch": 0.6265296133137542,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003195,
      "loss": 6.2341,
      "mean_token_accuracy": 0.14668046087026596,
      "num_tokens": 1526760.0,
      "step": 640
    },
    {
      "entropy": 6.560241603851319,
      "epoch": 0.631424375917768,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000322,
      "loss": 6.3118,
      "mean_token_accuracy": 0.14421986192464828,
      "num_tokens": 1537804.0,
      "step": 645
    },
    {
      "entropy": 6.549531269073486,
      "epoch": 0.6363191385217817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.3171,
      "mean_token_accuracy": 0.1333357997238636,
      "num_tokens": 1550423.0,
      "step": 650
    },
    {
      "entropy": 6.398512887954712,
      "epoch": 0.6412139011257953,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.218,
      "mean_token_accuracy": 0.1467081144452095,
      "num_tokens": 1563032.0,
      "step": 655
    },
    {
      "entropy": 6.56570143699646,
      "epoch": 0.6461086637298091,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.3105,
      "mean_token_accuracy": 0.13749707862734795,
      "num_tokens": 1575600.0,
      "step": 660
    },
    {
      "entropy": 6.476151990890503,
      "epoch": 0.6510034263338228,
      "grad_norm": 1.25,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.2174,
      "mean_token_accuracy": 0.1447758801281452,
      "num_tokens": 1587604.0,
      "step": 665
    },
    {
      "entropy": 6.376614093780518,
      "epoch": 0.6558981889378365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.24,
      "mean_token_accuracy": 0.13755062445998192,
      "num_tokens": 1599731.0,
      "step": 670
    },
    {
      "entropy": 6.469626808166504,
      "epoch": 0.6607929515418502,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000337,
      "loss": 6.282,
      "mean_token_accuracy": 0.1430655002593994,
      "num_tokens": 1611944.0,
      "step": 675
    },
    {
      "entropy": 6.544871091842651,
      "epoch": 0.6656877141458639,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003395,
      "loss": 6.1867,
      "mean_token_accuracy": 0.1425486721098423,
      "num_tokens": 1623177.0,
      "step": 680
    },
    {
      "entropy": 6.3893975734710695,
      "epoch": 0.6705824767498776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000342,
      "loss": 6.2379,
      "mean_token_accuracy": 0.14739967435598372,
      "num_tokens": 1635075.0,
      "step": 685
    },
    {
      "entropy": 6.408784580230713,
      "epoch": 0.6754772393538914,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.1784,
      "mean_token_accuracy": 0.14091105610132218,
      "num_tokens": 1645126.0,
      "step": 690
    },
    {
      "entropy": 6.488248634338379,
      "epoch": 0.680372001957905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000347,
      "loss": 6.2174,
      "mean_token_accuracy": 0.14224316030740738,
      "num_tokens": 1656009.0,
      "step": 695
    },
    {
      "entropy": 6.27689561843872,
      "epoch": 0.6852667645619187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003495,
      "loss": 6.1975,
      "mean_token_accuracy": 0.1452504150569439,
      "num_tokens": 1667546.0,
      "step": 700
    },
    {
      "entropy": 6.541668558120728,
      "epoch": 0.6901615271659325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000352,
      "loss": 6.252,
      "mean_token_accuracy": 0.13904105573892594,
      "num_tokens": 1679176.0,
      "step": 705
    },
    {
      "entropy": 6.330350160598755,
      "epoch": 0.6950562897699462,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003545,
      "loss": 6.1147,
      "mean_token_accuracy": 0.1483779713511467,
      "num_tokens": 1691240.0,
      "step": 710
    },
    {
      "entropy": 6.385787868499756,
      "epoch": 0.6999510523739598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000357,
      "loss": 6.1505,
      "mean_token_accuracy": 0.1480160176753998,
      "num_tokens": 1702185.0,
      "step": 715
    },
    {
      "entropy": 6.384578990936279,
      "epoch": 0.7048458149779736,
      "grad_norm": 1.25,
      "learning_rate": 0.0003595,
      "loss": 6.1536,
      "mean_token_accuracy": 0.14635321646928787,
      "num_tokens": 1713702.0,
      "step": 720
    },
    {
      "entropy": 6.361242246627808,
      "epoch": 0.7097405775819873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000362,
      "loss": 6.2363,
      "mean_token_accuracy": 0.14713326394557952,
      "num_tokens": 1724943.0,
      "step": 725
    },
    {
      "entropy": 6.3330967903137205,
      "epoch": 0.714635340186001,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003645,
      "loss": 6.2454,
      "mean_token_accuracy": 0.1357298105955124,
      "num_tokens": 1737055.0,
      "step": 730
    },
    {
      "entropy": 6.478425121307373,
      "epoch": 0.7195301027900147,
      "grad_norm": 1.25,
      "learning_rate": 0.000367,
      "loss": 6.2552,
      "mean_token_accuracy": 0.13602484837174417,
      "num_tokens": 1749772.0,
      "step": 735
    },
    {
      "entropy": 6.379587030410766,
      "epoch": 0.7244248653940284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003695,
      "loss": 6.2908,
      "mean_token_accuracy": 0.13715352192521096,
      "num_tokens": 1762533.0,
      "step": 740
    },
    {
      "entropy": 6.489045238494873,
      "epoch": 0.7293196279980421,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000372,
      "loss": 6.2885,
      "mean_token_accuracy": 0.14545208886265754,
      "num_tokens": 1774541.0,
      "step": 745
    },
    {
      "entropy": 6.449005842208862,
      "epoch": 0.7342143906020558,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003745,
      "loss": 6.2495,
      "mean_token_accuracy": 0.14404454231262206,
      "num_tokens": 1785627.0,
      "step": 750
    },
    {
      "entropy": 6.345538234710693,
      "epoch": 0.7391091532060695,
      "grad_norm": 1.21875,
      "learning_rate": 0.000377,
      "loss": 6.074,
      "mean_token_accuracy": 0.15339891761541366,
      "num_tokens": 1796106.0,
      "step": 755
    },
    {
      "entropy": 6.269445323944092,
      "epoch": 0.7440039158100832,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003795,
      "loss": 6.1115,
      "mean_token_accuracy": 0.1419329449534416,
      "num_tokens": 1808228.0,
      "step": 760
    },
    {
      "entropy": 6.318842649459839,
      "epoch": 0.748898678414097,
      "grad_norm": 1.34375,
      "learning_rate": 0.000382,
      "loss": 6.1434,
      "mean_token_accuracy": 0.1400843970477581,
      "num_tokens": 1819415.0,
      "step": 765
    },
    {
      "entropy": 6.33818244934082,
      "epoch": 0.7537934410181106,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003845,
      "loss": 6.1994,
      "mean_token_accuracy": 0.14103697687387468,
      "num_tokens": 1831043.0,
      "step": 770
    },
    {
      "entropy": 6.344356966018677,
      "epoch": 0.7586882036221243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.1764,
      "mean_token_accuracy": 0.13956040516495705,
      "num_tokens": 1843062.0,
      "step": 775
    },
    {
      "entropy": 6.430766439437866,
      "epoch": 0.7635829662261381,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.2983,
      "mean_token_accuracy": 0.1392583154141903,
      "num_tokens": 1855181.0,
      "step": 780
    },
    {
      "entropy": 6.23683648109436,
      "epoch": 0.7684777288301518,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.0649,
      "mean_token_accuracy": 0.147367200255394,
      "num_tokens": 1866626.0,
      "step": 785
    },
    {
      "entropy": 6.35269775390625,
      "epoch": 0.7733724914341654,
      "grad_norm": 1.375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.2047,
      "mean_token_accuracy": 0.1448512777686119,
      "num_tokens": 1878887.0,
      "step": 790
    },
    {
      "entropy": 6.355936002731323,
      "epoch": 0.7782672540381792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.1022,
      "mean_token_accuracy": 0.14598082453012468,
      "num_tokens": 1890592.0,
      "step": 795
    },
    {
      "entropy": 6.235448217391967,
      "epoch": 0.7831620166421929,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003995,
      "loss": 6.142,
      "mean_token_accuracy": 0.14651560485363008,
      "num_tokens": 1901875.0,
      "step": 800
    },
    {
      "entropy": 6.273421049118042,
      "epoch": 0.7880567792462065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000402,
      "loss": 6.1474,
      "mean_token_accuracy": 0.13966845646500586,
      "num_tokens": 1913913.0,
      "step": 805
    },
    {
      "entropy": 6.30178804397583,
      "epoch": 0.7929515418502202,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004045,
      "loss": 6.0861,
      "mean_token_accuracy": 0.1498040184378624,
      "num_tokens": 1924667.0,
      "step": 810
    },
    {
      "entropy": 6.291550588607788,
      "epoch": 0.797846304454234,
      "grad_norm": 1.25,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.1604,
      "mean_token_accuracy": 0.14530218094587327,
      "num_tokens": 1935838.0,
      "step": 815
    },
    {
      "entropy": 6.291401100158692,
      "epoch": 0.8027410670582477,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004095,
      "loss": 6.1019,
      "mean_token_accuracy": 0.15018822252750397,
      "num_tokens": 1947608.0,
      "step": 820
    },
    {
      "entropy": 6.300445222854615,
      "epoch": 0.8076358296622613,
      "grad_norm": 1.34375,
      "learning_rate": 0.000412,
      "loss": 6.0924,
      "mean_token_accuracy": 0.1496323451399803,
      "num_tokens": 1958426.0,
      "step": 825
    },
    {
      "entropy": 6.2596992492675785,
      "epoch": 0.8125305922662751,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004145,
      "loss": 6.1174,
      "mean_token_accuracy": 0.14941085800528525,
      "num_tokens": 1971193.0,
      "step": 830
    },
    {
      "entropy": 6.247623252868652,
      "epoch": 0.8174253548702888,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000417,
      "loss": 6.0576,
      "mean_token_accuracy": 0.14635169357061387,
      "num_tokens": 1982019.0,
      "step": 835
    },
    {
      "entropy": 6.237645626068115,
      "epoch": 0.8223201174743024,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004195,
      "loss": 6.1127,
      "mean_token_accuracy": 0.15242150872945787,
      "num_tokens": 1993991.0,
      "step": 840
    },
    {
      "entropy": 6.244370412826538,
      "epoch": 0.8272148800783162,
      "grad_norm": 1.078125,
      "learning_rate": 0.000422,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14876028150320053,
      "num_tokens": 2006264.0,
      "step": 845
    },
    {
      "entropy": 6.165139627456665,
      "epoch": 0.8321096426823299,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004245,
      "loss": 6.1227,
      "mean_token_accuracy": 0.15057290643453597,
      "num_tokens": 2018308.0,
      "step": 850
    },
    {
      "entropy": 6.253602123260498,
      "epoch": 0.8370044052863436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000427,
      "loss": 6.0878,
      "mean_token_accuracy": 0.14493273943662643,
      "num_tokens": 2030829.0,
      "step": 855
    },
    {
      "entropy": 6.2643944263458256,
      "epoch": 0.8418991678903573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004295,
      "loss": 6.1606,
      "mean_token_accuracy": 0.14317945316433905,
      "num_tokens": 2041897.0,
      "step": 860
    },
    {
      "entropy": 6.1602997303009035,
      "epoch": 0.846793930494371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000432,
      "loss": 6.0424,
      "mean_token_accuracy": 0.15380731970071793,
      "num_tokens": 2054279.0,
      "step": 865
    },
    {
      "entropy": 6.201865339279175,
      "epoch": 0.8516886930983847,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004345,
      "loss": 6.1401,
      "mean_token_accuracy": 0.1524582713842392,
      "num_tokens": 2066150.0,
      "step": 870
    },
    {
      "entropy": 6.292105579376221,
      "epoch": 0.8565834557023985,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 6.1618,
      "mean_token_accuracy": 0.1416109673678875,
      "num_tokens": 2077895.0,
      "step": 875
    },
    {
      "entropy": 6.222372436523438,
      "epoch": 0.8614782183064121,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 6.0908,
      "mean_token_accuracy": 0.1596553936600685,
      "num_tokens": 2089205.0,
      "step": 880
    },
    {
      "entropy": 6.336112594604492,
      "epoch": 0.8663729809104258,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000442,
      "loss": 6.187,
      "mean_token_accuracy": 0.13830628544092177,
      "num_tokens": 2099939.0,
      "step": 885
    },
    {
      "entropy": 6.246317911148071,
      "epoch": 0.8712677435144396,
      "grad_norm": 1.25,
      "learning_rate": 0.0004445,
      "loss": 6.171,
      "mean_token_accuracy": 0.1431927777826786,
      "num_tokens": 2112148.0,
      "step": 890
    },
    {
      "entropy": 6.198504447937012,
      "epoch": 0.8761625061184533,
      "grad_norm": 1.390625,
      "learning_rate": 0.000447,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1506725162267685,
      "num_tokens": 2123375.0,
      "step": 895
    },
    {
      "entropy": 6.207621812820435,
      "epoch": 0.8810572687224669,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14134914726018905,
      "num_tokens": 2135884.0,
      "step": 900
    },
    {
      "entropy": 6.248273086547852,
      "epoch": 0.8859520313264807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.1078,
      "mean_token_accuracy": 0.14830219596624375,
      "num_tokens": 2147912.0,
      "step": 905
    },
    {
      "entropy": 6.193967342376709,
      "epoch": 0.8908467939304944,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.0775,
      "mean_token_accuracy": 0.1499609798192978,
      "num_tokens": 2159683.0,
      "step": 910
    },
    {
      "entropy": 6.1300129890441895,
      "epoch": 0.895741556534508,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045700000000000005,
      "loss": 5.9946,
      "mean_token_accuracy": 0.1463859647512436,
      "num_tokens": 2172528.0,
      "step": 915
    },
    {
      "entropy": 6.162504768371582,
      "epoch": 0.9006363191385218,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045950000000000006,
      "loss": 5.9592,
      "mean_token_accuracy": 0.15458274483680726,
      "num_tokens": 2183947.0,
      "step": 920
    },
    {
      "entropy": 6.154705381393432,
      "epoch": 0.9055310817425355,
      "grad_norm": 1.3125,
      "learning_rate": 0.000462,
      "loss": 6.0676,
      "mean_token_accuracy": 0.1460274837911129,
      "num_tokens": 2195167.0,
      "step": 925
    },
    {
      "entropy": 6.221498584747314,
      "epoch": 0.9104258443465492,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004645,
      "loss": 6.0434,
      "mean_token_accuracy": 0.1462668165564537,
      "num_tokens": 2207830.0,
      "step": 930
    },
    {
      "entropy": 6.170248174667359,
      "epoch": 0.9153206069505629,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000467,
      "loss": 6.0834,
      "mean_token_accuracy": 0.15441582649946212,
      "num_tokens": 2220717.0,
      "step": 935
    },
    {
      "entropy": 6.178001356124878,
      "epoch": 0.9202153695545766,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004695,
      "loss": 6.2167,
      "mean_token_accuracy": 0.14381293207406998,
      "num_tokens": 2232789.0,
      "step": 940
    },
    {
      "entropy": 6.206983280181885,
      "epoch": 0.9251101321585903,
      "grad_norm": 1.1875,
      "learning_rate": 0.000472,
      "loss": 6.0395,
      "mean_token_accuracy": 0.14444569125771523,
      "num_tokens": 2245155.0,
      "step": 945
    },
    {
      "entropy": 6.121549892425537,
      "epoch": 0.9300048947626041,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004745,
      "loss": 6.012,
      "mean_token_accuracy": 0.15058331415057183,
      "num_tokens": 2256951.0,
      "step": 950
    },
    {
      "entropy": 6.0408790588378904,
      "epoch": 0.9348996573666177,
      "grad_norm": 1.0625,
      "learning_rate": 0.000477,
      "loss": 6.0043,
      "mean_token_accuracy": 0.15417952388525008,
      "num_tokens": 2269087.0,
      "step": 955
    },
    {
      "entropy": 6.1350022315979,
      "epoch": 0.9397944199706314,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004795,
      "loss": 6.0201,
      "mean_token_accuracy": 0.1560334235429764,
      "num_tokens": 2280563.0,
      "step": 960
    },
    {
      "entropy": 6.190943622589112,
      "epoch": 0.9446891825746452,
      "grad_norm": 1.125,
      "learning_rate": 0.000482,
      "loss": 6.2106,
      "mean_token_accuracy": 0.14610329791903495,
      "num_tokens": 2292454.0,
      "step": 965
    },
    {
      "entropy": 6.0746870040893555,
      "epoch": 0.9495839451786589,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004845,
      "loss": 5.9563,
      "mean_token_accuracy": 0.15315437838435172,
      "num_tokens": 2305414.0,
      "step": 970
    },
    {
      "entropy": 6.177819538116455,
      "epoch": 0.9544787077826725,
      "grad_norm": 1.125,
      "learning_rate": 0.000487,
      "loss": 6.0325,
      "mean_token_accuracy": 0.1509896457195282,
      "num_tokens": 2316720.0,
      "step": 975
    },
    {
      "entropy": 6.1112165451049805,
      "epoch": 0.9593734703866863,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895,
      "loss": 6.0576,
      "mean_token_accuracy": 0.1500990182161331,
      "num_tokens": 2330077.0,
      "step": 980
    },
    {
      "entropy": 6.1100501537323,
      "epoch": 0.9642682329907,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492,
      "loss": 6.0185,
      "mean_token_accuracy": 0.14640207290649415,
      "num_tokens": 2342943.0,
      "step": 985
    },
    {
      "entropy": 6.120457935333252,
      "epoch": 0.9691629955947136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004945,
      "loss": 6.0579,
      "mean_token_accuracy": 0.15151562839746474,
      "num_tokens": 2354770.0,
      "step": 990
    },
    {
      "entropy": 6.153151655197144,
      "epoch": 0.9740577581987273,
      "grad_norm": 1.125,
      "learning_rate": 0.000497,
      "loss": 6.0659,
      "mean_token_accuracy": 0.1489930845797062,
      "num_tokens": 2367061.0,
      "step": 995
    },
    {
      "entropy": 6.129442977905273,
      "epoch": 0.9789525208027411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995,
      "loss": 5.9541,
      "mean_token_accuracy": 0.1546485498547554,
      "num_tokens": 2379597.0,
      "step": 1000
    },
    {
      "epoch": 0.9789525208027411,
      "eval_entropy": 5.923030702807787,
      "eval_loss": 6.252185821533203,
      "eval_mean_token_accuracy": 0.14683223417214616,
      "eval_num_tokens": 2379597.0,
      "eval_runtime": 2.8719,
      "eval_samples_per_second": 1200.257,
      "eval_steps_per_second": 150.076,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 10210,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3688675946496000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}