{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.1478298238074776,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.69186782836914,
      "epoch": 0.004297378599054577,
      "grad_norm": 14.0625,
      "learning_rate": 2e-06,
      "loss": 10.767,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8559.0,
      "step": 5
    },
    {
      "entropy": 10.691853904724121,
      "epoch": 0.008594757198109154,
      "grad_norm": 13.4375,
      "learning_rate": 4.5e-06,
      "loss": 10.7343,
      "mean_token_accuracy": 0.0006453584705013782,
      "num_tokens": 18235.0,
      "step": 10
    },
    {
      "entropy": 10.68972110748291,
      "epoch": 0.01289213579716373,
      "grad_norm": 10.375,
      "learning_rate": 7e-06,
      "loss": 10.4481,
      "mean_token_accuracy": 0.04634292460978031,
      "num_tokens": 28257.0,
      "step": 15
    },
    {
      "entropy": 10.653952026367188,
      "epoch": 0.017189514396218308,
      "grad_norm": 6.28125,
      "learning_rate": 9.5e-06,
      "loss": 10.1016,
      "mean_token_accuracy": 0.05583016015589237,
      "num_tokens": 38142.0,
      "step": 20
    },
    {
      "entropy": 10.492306423187255,
      "epoch": 0.021486892995272882,
      "grad_norm": 3.90625,
      "learning_rate": 1.2e-05,
      "loss": 9.8388,
      "mean_token_accuracy": 0.051388037577271464,
      "num_tokens": 47363.0,
      "step": 25
    },
    {
      "entropy": 10.393956565856934,
      "epoch": 0.02578427159432746,
      "grad_norm": 3.09375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.6855,
      "mean_token_accuracy": 0.05032752677798271,
      "num_tokens": 57121.0,
      "step": 30
    },
    {
      "entropy": 10.352927017211915,
      "epoch": 0.030081650193382038,
      "grad_norm": 2.71875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.59,
      "mean_token_accuracy": 0.056775762140750884,
      "num_tokens": 66785.0,
      "step": 35
    },
    {
      "entropy": 10.365454959869385,
      "epoch": 0.034379028792436615,
      "grad_norm": 2.671875,
      "learning_rate": 1.95e-05,
      "loss": 9.4974,
      "mean_token_accuracy": 0.05595241449773312,
      "num_tokens": 76313.0,
      "step": 40
    },
    {
      "entropy": 10.268372726440429,
      "epoch": 0.03867640739149119,
      "grad_norm": 2.25,
      "learning_rate": 2.2e-05,
      "loss": 9.546,
      "mean_token_accuracy": 0.05818333774805069,
      "num_tokens": 86151.0,
      "step": 45
    },
    {
      "entropy": 10.402025985717774,
      "epoch": 0.042973785990545764,
      "grad_norm": 2.484375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4872,
      "mean_token_accuracy": 0.06535410135984421,
      "num_tokens": 94916.0,
      "step": 50
    },
    {
      "entropy": 10.304343509674073,
      "epoch": 0.047271164589600345,
      "grad_norm": 2.265625,
      "learning_rate": 2.7e-05,
      "loss": 9.4098,
      "mean_token_accuracy": 0.06619026586413383,
      "num_tokens": 104592.0,
      "step": 55
    },
    {
      "entropy": 10.089207649230957,
      "epoch": 0.05156854318865492,
      "grad_norm": 2.28125,
      "learning_rate": 2.95e-05,
      "loss": 9.317,
      "mean_token_accuracy": 0.07636640071868897,
      "num_tokens": 113942.0,
      "step": 60
    },
    {
      "entropy": 10.178072547912597,
      "epoch": 0.055865921787709494,
      "grad_norm": 2.6875,
      "learning_rate": 3.2e-05,
      "loss": 9.329,
      "mean_token_accuracy": 0.07130967006087303,
      "num_tokens": 123190.0,
      "step": 65
    },
    {
      "entropy": 10.223089408874511,
      "epoch": 0.060163300386764075,
      "grad_norm": 1.8125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3231,
      "mean_token_accuracy": 0.06454960964620113,
      "num_tokens": 132367.0,
      "step": 70
    },
    {
      "entropy": 10.157601451873779,
      "epoch": 0.06446067898581866,
      "grad_norm": 1.8828125,
      "learning_rate": 3.7e-05,
      "loss": 9.1389,
      "mean_token_accuracy": 0.07756163887679576,
      "num_tokens": 141375.0,
      "step": 75
    },
    {
      "entropy": 10.095236587524415,
      "epoch": 0.06875805758487323,
      "grad_norm": 1.859375,
      "learning_rate": 3.95e-05,
      "loss": 9.0625,
      "mean_token_accuracy": 0.07905065715312957,
      "num_tokens": 150791.0,
      "step": 80
    },
    {
      "entropy": 10.089058017730713,
      "epoch": 0.0730554361839278,
      "grad_norm": 1.96875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0294,
      "mean_token_accuracy": 0.07317423708736896,
      "num_tokens": 158686.0,
      "step": 85
    },
    {
      "entropy": 10.112849235534668,
      "epoch": 0.07735281478298238,
      "grad_norm": 1.65625,
      "learning_rate": 4.45e-05,
      "loss": 8.9757,
      "mean_token_accuracy": 0.06819208525121212,
      "num_tokens": 167482.0,
      "step": 90
    },
    {
      "entropy": 9.973779201507568,
      "epoch": 0.08165019338203695,
      "grad_norm": 2.046875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.802,
      "mean_token_accuracy": 0.08643188402056694,
      "num_tokens": 176598.0,
      "step": 95
    },
    {
      "entropy": 9.932288265228271,
      "epoch": 0.08594757198109153,
      "grad_norm": 1.8984375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7014,
      "mean_token_accuracy": 0.08371126018464566,
      "num_tokens": 185883.0,
      "step": 100
    },
    {
      "entropy": 9.903890705108642,
      "epoch": 0.09024495058014612,
      "grad_norm": 1.65625,
      "learning_rate": 5.2e-05,
      "loss": 8.672,
      "mean_token_accuracy": 0.0767141304910183,
      "num_tokens": 194820.0,
      "step": 105
    },
    {
      "entropy": 9.905889225006103,
      "epoch": 0.09454232917920069,
      "grad_norm": 1.8671875,
      "learning_rate": 5.45e-05,
      "loss": 8.5994,
      "mean_token_accuracy": 0.07663930654525757,
      "num_tokens": 203710.0,
      "step": 110
    },
    {
      "entropy": 9.7015700340271,
      "epoch": 0.09883970777825526,
      "grad_norm": 1.640625,
      "learning_rate": 5.7e-05,
      "loss": 8.4883,
      "mean_token_accuracy": 0.08735396713018417,
      "num_tokens": 212642.0,
      "step": 115
    },
    {
      "entropy": 9.712564849853516,
      "epoch": 0.10313708637730984,
      "grad_norm": 1.4375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.4829,
      "mean_token_accuracy": 0.07560812048614025,
      "num_tokens": 221411.0,
      "step": 120
    },
    {
      "entropy": 9.618709659576416,
      "epoch": 0.10743446497636441,
      "grad_norm": 1.3828125,
      "learning_rate": 6.2e-05,
      "loss": 8.276,
      "mean_token_accuracy": 0.07886423245072365,
      "num_tokens": 230454.0,
      "step": 125
    },
    {
      "entropy": 9.445035266876221,
      "epoch": 0.11173184357541899,
      "grad_norm": 1.78125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.2553,
      "mean_token_accuracy": 0.08224238455295563,
      "num_tokens": 240272.0,
      "step": 130
    },
    {
      "entropy": 9.343283653259277,
      "epoch": 0.11602922217447358,
      "grad_norm": 1.2109375,
      "learning_rate": 6.7e-05,
      "loss": 8.1773,
      "mean_token_accuracy": 0.07860668338835239,
      "num_tokens": 249652.0,
      "step": 135
    },
    {
      "entropy": 9.141777420043946,
      "epoch": 0.12032660077352815,
      "grad_norm": 1.359375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.0586,
      "mean_token_accuracy": 0.08432014659047127,
      "num_tokens": 258359.0,
      "step": 140
    },
    {
      "entropy": 9.028361320495605,
      "epoch": 0.12462397937258272,
      "grad_norm": 1.9296875,
      "learning_rate": 7.2e-05,
      "loss": 7.965,
      "mean_token_accuracy": 0.08845296204090118,
      "num_tokens": 266863.0,
      "step": 145
    },
    {
      "entropy": 8.872007942199707,
      "epoch": 0.1289213579716373,
      "grad_norm": 1.703125,
      "learning_rate": 7.45e-05,
      "loss": 7.9485,
      "mean_token_accuracy": 0.08544271066784859,
      "num_tokens": 274694.0,
      "step": 150
    },
    {
      "entropy": 8.746376705169677,
      "epoch": 0.1332187365706919,
      "grad_norm": 1.0078125,
      "learning_rate": 7.7e-05,
      "loss": 7.998,
      "mean_token_accuracy": 0.07424286939203739,
      "num_tokens": 285038.0,
      "step": 155
    },
    {
      "entropy": 8.652623462677003,
      "epoch": 0.13751611516974646,
      "grad_norm": 1.34375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9434,
      "mean_token_accuracy": 0.07819288298487663,
      "num_tokens": 293451.0,
      "step": 160
    },
    {
      "entropy": 8.498311519622803,
      "epoch": 0.14181349376880104,
      "grad_norm": 1.171875,
      "learning_rate": 8.2e-05,
      "loss": 7.7897,
      "mean_token_accuracy": 0.08914674445986748,
      "num_tokens": 302184.0,
      "step": 165
    },
    {
      "entropy": 8.397662544250489,
      "epoch": 0.1461108723678556,
      "grad_norm": 1.2109375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8596,
      "mean_token_accuracy": 0.08310422226786614,
      "num_tokens": 311069.0,
      "step": 170
    },
    {
      "entropy": 8.409068965911866,
      "epoch": 0.15040825096691018,
      "grad_norm": 1.25,
      "learning_rate": 8.7e-05,
      "loss": 7.8454,
      "mean_token_accuracy": 0.07755065970122814,
      "num_tokens": 319841.0,
      "step": 175
    },
    {
      "entropy": 8.261888885498047,
      "epoch": 0.15470562956596476,
      "grad_norm": 1.0546875,
      "learning_rate": 8.95e-05,
      "loss": 7.7995,
      "mean_token_accuracy": 0.0760527990758419,
      "num_tokens": 328621.0,
      "step": 180
    },
    {
      "entropy": 8.25572681427002,
      "epoch": 0.15900300816501933,
      "grad_norm": 1.46875,
      "learning_rate": 9.2e-05,
      "loss": 7.7913,
      "mean_token_accuracy": 0.07687021419405937,
      "num_tokens": 336469.0,
      "step": 185
    },
    {
      "entropy": 8.273696994781494,
      "epoch": 0.1633003867640739,
      "grad_norm": 1.21875,
      "learning_rate": 9.45e-05,
      "loss": 7.8148,
      "mean_token_accuracy": 0.07753114774823189,
      "num_tokens": 344859.0,
      "step": 190
    },
    {
      "entropy": 8.1952486038208,
      "epoch": 0.16759776536312848,
      "grad_norm": 1.0859375,
      "learning_rate": 9.7e-05,
      "loss": 7.8017,
      "mean_token_accuracy": 0.08184871897101402,
      "num_tokens": 354536.0,
      "step": 195
    },
    {
      "entropy": 8.130875205993652,
      "epoch": 0.17189514396218306,
      "grad_norm": 1.2265625,
      "learning_rate": 9.95e-05,
      "loss": 7.782,
      "mean_token_accuracy": 0.08327300399541855,
      "num_tokens": 363592.0,
      "step": 200
    },
    {
      "entropy": 8.071674633026124,
      "epoch": 0.17619252256123766,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000102,
      "loss": 7.7545,
      "mean_token_accuracy": 0.08299842327833176,
      "num_tokens": 373278.0,
      "step": 205
    },
    {
      "entropy": 8.152774953842163,
      "epoch": 0.18048990116029223,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.8002,
      "mean_token_accuracy": 0.08091530725359916,
      "num_tokens": 382498.0,
      "step": 210
    },
    {
      "entropy": 8.122348976135253,
      "epoch": 0.1847872797593468,
      "grad_norm": 1.046875,
      "learning_rate": 0.000107,
      "loss": 7.7393,
      "mean_token_accuracy": 0.09220762960612774,
      "num_tokens": 391273.0,
      "step": 215
    },
    {
      "entropy": 8.141310453414917,
      "epoch": 0.18908465835840138,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001095,
      "loss": 7.6711,
      "mean_token_accuracy": 0.07944022566080093,
      "num_tokens": 400340.0,
      "step": 220
    },
    {
      "entropy": 8.08761625289917,
      "epoch": 0.19338203695745596,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000112,
      "loss": 7.7203,
      "mean_token_accuracy": 0.07989789322018623,
      "num_tokens": 409162.0,
      "step": 225
    },
    {
      "entropy": 8.068865013122558,
      "epoch": 0.19767941555651053,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001145,
      "loss": 7.6955,
      "mean_token_accuracy": 0.08263300582766533,
      "num_tokens": 417345.0,
      "step": 230
    },
    {
      "entropy": 8.020773887634277,
      "epoch": 0.2019767941555651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.6885,
      "mean_token_accuracy": 0.08577688783407211,
      "num_tokens": 427488.0,
      "step": 235
    },
    {
      "entropy": 8.00288724899292,
      "epoch": 0.20627417275461968,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.6381,
      "mean_token_accuracy": 0.08082681894302368,
      "num_tokens": 435496.0,
      "step": 240
    },
    {
      "entropy": 8.024074983596801,
      "epoch": 0.21057155135367425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000122,
      "loss": 7.592,
      "mean_token_accuracy": 0.08299721553921699,
      "num_tokens": 444290.0,
      "step": 245
    },
    {
      "entropy": 8.002217769622803,
      "epoch": 0.21486892995272883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001245,
      "loss": 7.7047,
      "mean_token_accuracy": 0.0830993577837944,
      "num_tokens": 452704.0,
      "step": 250
    },
    {
      "entropy": 8.036895895004273,
      "epoch": 0.2191663085517834,
      "grad_norm": 1.40625,
      "learning_rate": 0.000127,
      "loss": 7.6246,
      "mean_token_accuracy": 0.08470720425248146,
      "num_tokens": 461445.0,
      "step": 255
    },
    {
      "entropy": 8.038204240798951,
      "epoch": 0.22346368715083798,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001295,
      "loss": 7.6401,
      "mean_token_accuracy": 0.08966793864965439,
      "num_tokens": 470644.0,
      "step": 260
    },
    {
      "entropy": 7.892484998703003,
      "epoch": 0.22776106574989258,
      "grad_norm": 1.671875,
      "learning_rate": 0.000132,
      "loss": 7.5766,
      "mean_token_accuracy": 0.08981397226452828,
      "num_tokens": 479212.0,
      "step": 265
    },
    {
      "entropy": 7.9553313732147215,
      "epoch": 0.23205844434894715,
      "grad_norm": 1.09375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.6849,
      "mean_token_accuracy": 0.07991519421339036,
      "num_tokens": 489027.0,
      "step": 270
    },
    {
      "entropy": 8.001228666305542,
      "epoch": 0.23635582294800173,
      "grad_norm": 1.125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.7078,
      "mean_token_accuracy": 0.0864037737250328,
      "num_tokens": 497548.0,
      "step": 275
    },
    {
      "entropy": 7.95409255027771,
      "epoch": 0.2406532015470563,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001395,
      "loss": 7.523,
      "mean_token_accuracy": 0.09064189270138741,
      "num_tokens": 505836.0,
      "step": 280
    },
    {
      "entropy": 8.008459424972534,
      "epoch": 0.24495058014611087,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.6857,
      "mean_token_accuracy": 0.08223827481269837,
      "num_tokens": 515463.0,
      "step": 285
    },
    {
      "entropy": 8.042573165893554,
      "epoch": 0.24924795874516545,
      "grad_norm": 1.25,
      "learning_rate": 0.0001445,
      "loss": 7.5642,
      "mean_token_accuracy": 0.08510619327425957,
      "num_tokens": 524882.0,
      "step": 290
    },
    {
      "entropy": 7.878521060943603,
      "epoch": 0.25354533734422,
      "grad_norm": 1.171875,
      "learning_rate": 0.000147,
      "loss": 7.5627,
      "mean_token_accuracy": 0.08412178531289101,
      "num_tokens": 534523.0,
      "step": 295
    },
    {
      "entropy": 7.9167151927948,
      "epoch": 0.2578427159432746,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001495,
      "loss": 7.5608,
      "mean_token_accuracy": 0.08494351357221604,
      "num_tokens": 543772.0,
      "step": 300
    },
    {
      "entropy": 7.85640025138855,
      "epoch": 0.26214009454232917,
      "grad_norm": 1.203125,
      "learning_rate": 0.000152,
      "loss": 7.5374,
      "mean_token_accuracy": 0.0830413281917572,
      "num_tokens": 552024.0,
      "step": 305
    },
    {
      "entropy": 7.888205242156983,
      "epoch": 0.2664374731413838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.5269,
      "mean_token_accuracy": 0.08928481489419937,
      "num_tokens": 561114.0,
      "step": 310
    },
    {
      "entropy": 8.020092916488647,
      "epoch": 0.2707348517404383,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000157,
      "loss": 7.5728,
      "mean_token_accuracy": 0.08862031325697899,
      "num_tokens": 569655.0,
      "step": 315
    },
    {
      "entropy": 7.867624950408936,
      "epoch": 0.2750322303394929,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001595,
      "loss": 7.5604,
      "mean_token_accuracy": 0.08833226263523102,
      "num_tokens": 578701.0,
      "step": 320
    },
    {
      "entropy": 7.848216819763183,
      "epoch": 0.27932960893854747,
      "grad_norm": 1.25,
      "learning_rate": 0.000162,
      "loss": 7.5257,
      "mean_token_accuracy": 0.08904304206371308,
      "num_tokens": 588166.0,
      "step": 325
    },
    {
      "entropy": 7.775031900405883,
      "epoch": 0.28362698753760207,
      "grad_norm": 1.125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.5406,
      "mean_token_accuracy": 0.08560147807002068,
      "num_tokens": 598862.0,
      "step": 330
    },
    {
      "entropy": 7.946314001083374,
      "epoch": 0.2879243661366566,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.5054,
      "mean_token_accuracy": 0.08752046972513199,
      "num_tokens": 607680.0,
      "step": 335
    },
    {
      "entropy": 7.798824310302734,
      "epoch": 0.2922217447357112,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.5409,
      "mean_token_accuracy": 0.08527569100260735,
      "num_tokens": 617194.0,
      "step": 340
    },
    {
      "entropy": 7.840095901489258,
      "epoch": 0.29651912333476577,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.4681,
      "mean_token_accuracy": 0.09199495688080787,
      "num_tokens": 625946.0,
      "step": 345
    },
    {
      "entropy": 7.863882684707642,
      "epoch": 0.30081650193382037,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.4424,
      "mean_token_accuracy": 0.09303685426712036,
      "num_tokens": 634293.0,
      "step": 350
    },
    {
      "entropy": 7.814202308654785,
      "epoch": 0.30511388053287497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000177,
      "loss": 7.4365,
      "mean_token_accuracy": 0.08837181255221367,
      "num_tokens": 643479.0,
      "step": 355
    },
    {
      "entropy": 7.780666065216065,
      "epoch": 0.3094112591319295,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001795,
      "loss": 7.4565,
      "mean_token_accuracy": 0.0916847325861454,
      "num_tokens": 653599.0,
      "step": 360
    },
    {
      "entropy": 7.727724599838257,
      "epoch": 0.3137086377309841,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000182,
      "loss": 7.4396,
      "mean_token_accuracy": 0.0935065358877182,
      "num_tokens": 663042.0,
      "step": 365
    },
    {
      "entropy": 7.7599810600280765,
      "epoch": 0.31800601633003867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001845,
      "loss": 7.3799,
      "mean_token_accuracy": 0.09194350391626357,
      "num_tokens": 672254.0,
      "step": 370
    },
    {
      "entropy": 7.850320053100586,
      "epoch": 0.32230339492909327,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000187,
      "loss": 7.4208,
      "mean_token_accuracy": 0.09203241541981697,
      "num_tokens": 681648.0,
      "step": 375
    },
    {
      "entropy": 7.66558690071106,
      "epoch": 0.3266007735281478,
      "grad_norm": 1.25,
      "learning_rate": 0.0001895,
      "loss": 7.3756,
      "mean_token_accuracy": 0.09696918800473213,
      "num_tokens": 691080.0,
      "step": 380
    },
    {
      "entropy": 7.681433391571045,
      "epoch": 0.3308981521272024,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000192,
      "loss": 7.4656,
      "mean_token_accuracy": 0.08918402977287769,
      "num_tokens": 700549.0,
      "step": 385
    },
    {
      "entropy": 7.880253696441651,
      "epoch": 0.33519553072625696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001945,
      "loss": 7.5014,
      "mean_token_accuracy": 0.09039791896939278,
      "num_tokens": 710623.0,
      "step": 390
    },
    {
      "entropy": 7.726108407974243,
      "epoch": 0.33949290932531156,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.4269,
      "mean_token_accuracy": 0.09598973840475082,
      "num_tokens": 719695.0,
      "step": 395
    },
    {
      "entropy": 7.658704042434692,
      "epoch": 0.3437902879243661,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.3817,
      "mean_token_accuracy": 0.09576035216450692,
      "num_tokens": 728998.0,
      "step": 400
    },
    {
      "entropy": 7.817303800582886,
      "epoch": 0.3480876665234207,
      "grad_norm": 1.25,
      "learning_rate": 0.000202,
      "loss": 7.4597,
      "mean_token_accuracy": 0.08954662233591079,
      "num_tokens": 738537.0,
      "step": 405
    },
    {
      "entropy": 7.673436260223388,
      "epoch": 0.3523850451224753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.4039,
      "mean_token_accuracy": 0.092615757137537,
      "num_tokens": 747555.0,
      "step": 410
    },
    {
      "entropy": 7.643343782424926,
      "epoch": 0.35668242372152986,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000207,
      "loss": 7.3747,
      "mean_token_accuracy": 0.09669034704566001,
      "num_tokens": 756857.0,
      "step": 415
    },
    {
      "entropy": 7.74475588798523,
      "epoch": 0.36097980232058446,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002095,
      "loss": 7.3802,
      "mean_token_accuracy": 0.10021467804908753,
      "num_tokens": 766181.0,
      "step": 420
    },
    {
      "entropy": 7.649684762954712,
      "epoch": 0.365277180919639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000212,
      "loss": 7.4102,
      "mean_token_accuracy": 0.09460910633206368,
      "num_tokens": 774894.0,
      "step": 425
    },
    {
      "entropy": 7.730149173736573,
      "epoch": 0.3695745595186936,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002145,
      "loss": 7.4062,
      "mean_token_accuracy": 0.090732641518116,
      "num_tokens": 784601.0,
      "step": 430
    },
    {
      "entropy": 7.6217389583587645,
      "epoch": 0.37387193811774816,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.4417,
      "mean_token_accuracy": 0.09168312624096871,
      "num_tokens": 794373.0,
      "step": 435
    },
    {
      "entropy": 7.600076055526733,
      "epoch": 0.37816931671680276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002195,
      "loss": 7.3455,
      "mean_token_accuracy": 0.09434943050146102,
      "num_tokens": 803608.0,
      "step": 440
    },
    {
      "entropy": 7.752554702758789,
      "epoch": 0.3824666953158573,
      "grad_norm": 1.203125,
      "learning_rate": 0.000222,
      "loss": 7.3726,
      "mean_token_accuracy": 0.09299151077866555,
      "num_tokens": 812287.0,
      "step": 445
    },
    {
      "entropy": 7.518798732757569,
      "epoch": 0.3867640739149119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002245,
      "loss": 7.2985,
      "mean_token_accuracy": 0.0987668864428997,
      "num_tokens": 821049.0,
      "step": 450
    },
    {
      "entropy": 7.701573610305786,
      "epoch": 0.39106145251396646,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.3033,
      "mean_token_accuracy": 0.10087452456355095,
      "num_tokens": 829148.0,
      "step": 455
    },
    {
      "entropy": 7.614718294143676,
      "epoch": 0.39535883111302106,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.338,
      "mean_token_accuracy": 0.09538405686616898,
      "num_tokens": 838228.0,
      "step": 460
    },
    {
      "entropy": 7.615123701095581,
      "epoch": 0.39965620971207566,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.3246,
      "mean_token_accuracy": 0.09597658589482308,
      "num_tokens": 846059.0,
      "step": 465
    },
    {
      "entropy": 7.705140447616577,
      "epoch": 0.4039535883111302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.4473,
      "mean_token_accuracy": 0.09011114612221718,
      "num_tokens": 856309.0,
      "step": 470
    },
    {
      "entropy": 7.4943718910217285,
      "epoch": 0.4082509669101848,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000237,
      "loss": 7.2909,
      "mean_token_accuracy": 0.1010310634970665,
      "num_tokens": 865921.0,
      "step": 475
    },
    {
      "entropy": 7.549269151687622,
      "epoch": 0.41254834550923936,
      "grad_norm": 1.125,
      "learning_rate": 0.0002395,
      "loss": 7.2654,
      "mean_token_accuracy": 0.10411950796842576,
      "num_tokens": 875076.0,
      "step": 480
    },
    {
      "entropy": 7.629985094070435,
      "epoch": 0.41684572410829396,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000242,
      "loss": 7.3165,
      "mean_token_accuracy": 0.0936420701444149,
      "num_tokens": 884586.0,
      "step": 485
    },
    {
      "entropy": 7.624928522109985,
      "epoch": 0.4211431027073485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002445,
      "loss": 7.4262,
      "mean_token_accuracy": 0.09749620333313942,
      "num_tokens": 894192.0,
      "step": 490
    },
    {
      "entropy": 7.579270219802856,
      "epoch": 0.4254404813064031,
      "grad_norm": 1.078125,
      "learning_rate": 0.000247,
      "loss": 7.3251,
      "mean_token_accuracy": 0.09591478407382965,
      "num_tokens": 903542.0,
      "step": 495
    },
    {
      "entropy": 7.532130336761474,
      "epoch": 0.42973785990545765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002495,
      "loss": 7.259,
      "mean_token_accuracy": 0.10112758129835128,
      "num_tokens": 912625.0,
      "step": 500
    },
    {
      "epoch": 0.42973785990545765,
      "eval_entropy": 7.499067693143277,
      "eval_loss": 7.302952289581299,
      "eval_mean_token_accuracy": 0.1013092825793334,
      "eval_num_tokens": 912625.0,
      "eval_runtime": 2.6557,
      "eval_samples_per_second": 1336.368,
      "eval_steps_per_second": 167.187,
      "step": 500
    },
    {
      "entropy": 7.4851487159729,
      "epoch": 0.43403523850451226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000252,
      "loss": 7.218,
      "mean_token_accuracy": 0.10128417685627937,
      "num_tokens": 921438.0,
      "step": 505
    },
    {
      "entropy": 7.5242674350738525,
      "epoch": 0.4383326171035668,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002545,
      "loss": 7.2178,
      "mean_token_accuracy": 0.10356576293706894,
      "num_tokens": 931097.0,
      "step": 510
    },
    {
      "entropy": 7.51934700012207,
      "epoch": 0.4426299957026214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 7.1837,
      "mean_token_accuracy": 0.10203342065215111,
      "num_tokens": 939375.0,
      "step": 515
    },
    {
      "entropy": 7.429151105880737,
      "epoch": 0.44692737430167595,
      "grad_norm": 1.625,
      "learning_rate": 0.0002595,
      "loss": 7.1856,
      "mean_token_accuracy": 0.10627232864499092,
      "num_tokens": 948223.0,
      "step": 520
    },
    {
      "entropy": 7.493127393722534,
      "epoch": 0.45122475290073055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000262,
      "loss": 7.2191,
      "mean_token_accuracy": 0.09962140917778015,
      "num_tokens": 956243.0,
      "step": 525
    },
    {
      "entropy": 7.528323316574097,
      "epoch": 0.45552213149978515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.2164,
      "mean_token_accuracy": 0.10636536180973052,
      "num_tokens": 964506.0,
      "step": 530
    },
    {
      "entropy": 7.4207946300506595,
      "epoch": 0.4598195100988397,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.2399,
      "mean_token_accuracy": 0.0981321543455124,
      "num_tokens": 973617.0,
      "step": 535
    },
    {
      "entropy": 7.458613157272339,
      "epoch": 0.4641168886978943,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.153,
      "mean_token_accuracy": 0.10467413961887359,
      "num_tokens": 981796.0,
      "step": 540
    },
    {
      "entropy": 7.461965847015381,
      "epoch": 0.46841426729694885,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.2143,
      "mean_token_accuracy": 0.10574938952922822,
      "num_tokens": 990677.0,
      "step": 545
    },
    {
      "entropy": 7.328928756713867,
      "epoch": 0.47271164589600345,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002745,
      "loss": 7.1479,
      "mean_token_accuracy": 0.10492404326796531,
      "num_tokens": 1001131.0,
      "step": 550
    },
    {
      "entropy": 7.405250549316406,
      "epoch": 0.477009024495058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000277,
      "loss": 7.2806,
      "mean_token_accuracy": 0.0978497438132763,
      "num_tokens": 1009925.0,
      "step": 555
    },
    {
      "entropy": 7.450163650512695,
      "epoch": 0.4813064030941126,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002795,
      "loss": 7.211,
      "mean_token_accuracy": 0.10683610290288925,
      "num_tokens": 1018998.0,
      "step": 560
    },
    {
      "entropy": 7.455226850509644,
      "epoch": 0.48560378169316715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.1873,
      "mean_token_accuracy": 0.10439918413758278,
      "num_tokens": 1027496.0,
      "step": 565
    },
    {
      "entropy": 7.419484996795655,
      "epoch": 0.48990116029222175,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002845,
      "loss": 7.1247,
      "mean_token_accuracy": 0.11058635860681534,
      "num_tokens": 1035669.0,
      "step": 570
    },
    {
      "entropy": 7.423511219024658,
      "epoch": 0.4941985388912763,
      "grad_norm": 1.171875,
      "learning_rate": 0.000287,
      "loss": 7.2032,
      "mean_token_accuracy": 0.10203403383493423,
      "num_tokens": 1044416.0,
      "step": 575
    },
    {
      "entropy": 7.437924432754516,
      "epoch": 0.4984959174903309,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002895,
      "loss": 7.2185,
      "mean_token_accuracy": 0.10163830146193505,
      "num_tokens": 1053500.0,
      "step": 580
    },
    {
      "entropy": 7.371481227874756,
      "epoch": 0.5027932960893855,
      "grad_norm": 1.15625,
      "learning_rate": 0.000292,
      "loss": 7.1261,
      "mean_token_accuracy": 0.10348242074251175,
      "num_tokens": 1061758.0,
      "step": 585
    },
    {
      "entropy": 7.363918590545654,
      "epoch": 0.50709067468844,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002945,
      "loss": 7.1857,
      "mean_token_accuracy": 0.10252994000911712,
      "num_tokens": 1070570.0,
      "step": 590
    },
    {
      "entropy": 7.363972854614258,
      "epoch": 0.5113880532874946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000297,
      "loss": 7.1311,
      "mean_token_accuracy": 0.1051255501806736,
      "num_tokens": 1080640.0,
      "step": 595
    },
    {
      "entropy": 7.389889287948608,
      "epoch": 0.5156854318865493,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002995,
      "loss": 7.1887,
      "mean_token_accuracy": 0.10394371747970581,
      "num_tokens": 1090680.0,
      "step": 600
    },
    {
      "entropy": 7.449554491043091,
      "epoch": 0.5199828104856038,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000302,
      "loss": 7.2423,
      "mean_token_accuracy": 0.0991745762526989,
      "num_tokens": 1100453.0,
      "step": 605
    },
    {
      "entropy": 7.254623937606811,
      "epoch": 0.5242801890846583,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003045,
      "loss": 7.0726,
      "mean_token_accuracy": 0.11043546348810196,
      "num_tokens": 1109085.0,
      "step": 610
    },
    {
      "entropy": 7.319861698150635,
      "epoch": 0.5285775676837129,
      "grad_norm": 1.21875,
      "learning_rate": 0.000307,
      "loss": 7.0439,
      "mean_token_accuracy": 0.1101709671318531,
      "num_tokens": 1118333.0,
      "step": 615
    },
    {
      "entropy": 7.288142681121826,
      "epoch": 0.5328749462827675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003095,
      "loss": 7.0361,
      "mean_token_accuracy": 0.11464432403445243,
      "num_tokens": 1127459.0,
      "step": 620
    },
    {
      "entropy": 7.3332359790802,
      "epoch": 0.5371723248818221,
      "grad_norm": 1.15625,
      "learning_rate": 0.000312,
      "loss": 7.1691,
      "mean_token_accuracy": 0.1035873532295227,
      "num_tokens": 1137071.0,
      "step": 625
    },
    {
      "entropy": 7.295236444473266,
      "epoch": 0.5414697034808766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003145,
      "loss": 7.1535,
      "mean_token_accuracy": 0.10194391012191772,
      "num_tokens": 1146856.0,
      "step": 630
    },
    {
      "entropy": 7.367660856246948,
      "epoch": 0.5457670820799312,
      "grad_norm": 1.3125,
      "learning_rate": 0.000317,
      "loss": 7.1897,
      "mean_token_accuracy": 0.10307556018233299,
      "num_tokens": 1155847.0,
      "step": 635
    },
    {
      "entropy": 7.313199043273926,
      "epoch": 0.5500644606789858,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003195,
      "loss": 7.1504,
      "mean_token_accuracy": 0.10430814623832703,
      "num_tokens": 1164922.0,
      "step": 640
    },
    {
      "entropy": 7.366647577285766,
      "epoch": 0.5543618392780404,
      "grad_norm": 1.25,
      "learning_rate": 0.000322,
      "loss": 7.1199,
      "mean_token_accuracy": 0.10304575935006141,
      "num_tokens": 1174389.0,
      "step": 645
    },
    {
      "entropy": 7.232296657562256,
      "epoch": 0.5586592178770949,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.1016,
      "mean_token_accuracy": 0.10503174364566803,
      "num_tokens": 1182959.0,
      "step": 650
    },
    {
      "entropy": 7.329212665557861,
      "epoch": 0.5629565964761496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.1193,
      "mean_token_accuracy": 0.10655850470066071,
      "num_tokens": 1191777.0,
      "step": 655
    },
    {
      "entropy": 7.235855960845948,
      "epoch": 0.5672539750752041,
      "grad_norm": 1.125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.0269,
      "mean_token_accuracy": 0.11145639792084694,
      "num_tokens": 1200658.0,
      "step": 660
    },
    {
      "entropy": 7.195139932632446,
      "epoch": 0.5715513536742587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.0138,
      "mean_token_accuracy": 0.11196710541844368,
      "num_tokens": 1209607.0,
      "step": 665
    },
    {
      "entropy": 7.403189420700073,
      "epoch": 0.5758487322733132,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.158,
      "mean_token_accuracy": 0.10276976600289345,
      "num_tokens": 1218146.0,
      "step": 670
    },
    {
      "entropy": 7.122986936569214,
      "epoch": 0.5801461108723679,
      "grad_norm": 1.234375,
      "learning_rate": 0.000337,
      "loss": 7.0265,
      "mean_token_accuracy": 0.11236484497785568,
      "num_tokens": 1226886.0,
      "step": 675
    },
    {
      "entropy": 7.288639545440674,
      "epoch": 0.5844434894714224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003395,
      "loss": 7.0344,
      "mean_token_accuracy": 0.11365340650081635,
      "num_tokens": 1235408.0,
      "step": 680
    },
    {
      "entropy": 7.159159898757935,
      "epoch": 0.588740868070477,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000342,
      "loss": 7.025,
      "mean_token_accuracy": 0.11552022844552993,
      "num_tokens": 1244643.0,
      "step": 685
    },
    {
      "entropy": 7.252866315841675,
      "epoch": 0.5930382466695315,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.0859,
      "mean_token_accuracy": 0.10846259817481041,
      "num_tokens": 1254145.0,
      "step": 690
    },
    {
      "entropy": 7.239425182342529,
      "epoch": 0.5973356252685862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000347,
      "loss": 7.1242,
      "mean_token_accuracy": 0.10573266297578812,
      "num_tokens": 1263395.0,
      "step": 695
    },
    {
      "entropy": 7.336990690231323,
      "epoch": 0.6016330038676407,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003495,
      "loss": 7.0478,
      "mean_token_accuracy": 0.11208071708679199,
      "num_tokens": 1272942.0,
      "step": 700
    },
    {
      "entropy": 7.102862453460693,
      "epoch": 0.6059303824666953,
      "grad_norm": 1.171875,
      "learning_rate": 0.000352,
      "loss": 6.9782,
      "mean_token_accuracy": 0.11494042649865151,
      "num_tokens": 1281481.0,
      "step": 705
    },
    {
      "entropy": 7.3168416023254395,
      "epoch": 0.6102277610657499,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003545,
      "loss": 7.1398,
      "mean_token_accuracy": 0.10162141174077988,
      "num_tokens": 1290455.0,
      "step": 710
    },
    {
      "entropy": 7.213319635391235,
      "epoch": 0.6145251396648045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000357,
      "loss": 7.0151,
      "mean_token_accuracy": 0.1105075441300869,
      "num_tokens": 1299241.0,
      "step": 715
    },
    {
      "entropy": 7.2681577682495115,
      "epoch": 0.618822518263859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595,
      "loss": 7.0957,
      "mean_token_accuracy": 0.10155534744262695,
      "num_tokens": 1308658.0,
      "step": 720
    },
    {
      "entropy": 7.2813207626342775,
      "epoch": 0.6231198968629136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000362,
      "loss": 7.0292,
      "mean_token_accuracy": 0.10810133516788482,
      "num_tokens": 1316987.0,
      "step": 725
    },
    {
      "entropy": 7.062563180923462,
      "epoch": 0.6274172754619682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003645,
      "loss": 6.984,
      "mean_token_accuracy": 0.10933754518628121,
      "num_tokens": 1325500.0,
      "step": 730
    },
    {
      "entropy": 7.164379596710205,
      "epoch": 0.6317146540610228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000367,
      "loss": 7.1053,
      "mean_token_accuracy": 0.11041567921638488,
      "num_tokens": 1335048.0,
      "step": 735
    },
    {
      "entropy": 7.225457286834716,
      "epoch": 0.6360120326600773,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003695,
      "loss": 7.0434,
      "mean_token_accuracy": 0.11181584149599075,
      "num_tokens": 1343879.0,
      "step": 740
    },
    {
      "entropy": 7.160865116119385,
      "epoch": 0.6403094112591319,
      "grad_norm": 1.1875,
      "learning_rate": 0.000372,
      "loss": 6.9566,
      "mean_token_accuracy": 0.11149153709411622,
      "num_tokens": 1353169.0,
      "step": 745
    },
    {
      "entropy": 7.17898383140564,
      "epoch": 0.6446067898581865,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003745,
      "loss": 7.0493,
      "mean_token_accuracy": 0.10884664580225945,
      "num_tokens": 1362912.0,
      "step": 750
    },
    {
      "entropy": 7.154080009460449,
      "epoch": 0.6489041684572411,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 7.053,
      "mean_token_accuracy": 0.1088360756635666,
      "num_tokens": 1372331.0,
      "step": 755
    },
    {
      "entropy": 7.138390731811524,
      "epoch": 0.6532015470562956,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003795,
      "loss": 6.9359,
      "mean_token_accuracy": 0.11739541292190551,
      "num_tokens": 1381060.0,
      "step": 760
    },
    {
      "entropy": 7.090542840957641,
      "epoch": 0.6574989256553503,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000382,
      "loss": 6.9446,
      "mean_token_accuracy": 0.12408584356307983,
      "num_tokens": 1388749.0,
      "step": 765
    },
    {
      "entropy": 7.075214385986328,
      "epoch": 0.6617963042544048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003845,
      "loss": 6.9269,
      "mean_token_accuracy": 0.11776903718709945,
      "num_tokens": 1398456.0,
      "step": 770
    },
    {
      "entropy": 7.124542617797852,
      "epoch": 0.6660936828534594,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.9221,
      "mean_token_accuracy": 0.1167554147541523,
      "num_tokens": 1407430.0,
      "step": 775
    },
    {
      "entropy": 7.009809112548828,
      "epoch": 0.6703910614525139,
      "grad_norm": 1.125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.0097,
      "mean_token_accuracy": 0.10950447097420693,
      "num_tokens": 1416631.0,
      "step": 780
    },
    {
      "entropy": 7.185323333740234,
      "epoch": 0.6746884400515686,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.9554,
      "mean_token_accuracy": 0.1113646723330021,
      "num_tokens": 1426124.0,
      "step": 785
    },
    {
      "entropy": 7.108248376846314,
      "epoch": 0.6789858186506231,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.0781,
      "mean_token_accuracy": 0.10988424271345139,
      "num_tokens": 1436510.0,
      "step": 790
    },
    {
      "entropy": 7.196552371978759,
      "epoch": 0.6832831972496777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.007,
      "mean_token_accuracy": 0.10850347205996513,
      "num_tokens": 1445954.0,
      "step": 795
    },
    {
      "entropy": 7.0683338165283205,
      "epoch": 0.6875805758487322,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003995,
      "loss": 7.0326,
      "mean_token_accuracy": 0.10642682984471322,
      "num_tokens": 1454981.0,
      "step": 800
    },
    {
      "entropy": 7.130348634719849,
      "epoch": 0.6918779544477869,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000402,
      "loss": 6.981,
      "mean_token_accuracy": 0.11366944611072541,
      "num_tokens": 1464904.0,
      "step": 805
    },
    {
      "entropy": 7.058192777633667,
      "epoch": 0.6961753330468414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004045,
      "loss": 6.8463,
      "mean_token_accuracy": 0.1203661397099495,
      "num_tokens": 1473605.0,
      "step": 810
    },
    {
      "entropy": 7.020400810241699,
      "epoch": 0.700472711645896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8966,
      "mean_token_accuracy": 0.12048028632998467,
      "num_tokens": 1482989.0,
      "step": 815
    },
    {
      "entropy": 7.021880626678467,
      "epoch": 0.7047700902449506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004095,
      "loss": 6.9765,
      "mean_token_accuracy": 0.11234921589493752,
      "num_tokens": 1492514.0,
      "step": 820
    },
    {
      "entropy": 7.082749080657959,
      "epoch": 0.7090674688440052,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000412,
      "loss": 6.9776,
      "mean_token_accuracy": 0.11630082428455353,
      "num_tokens": 1501127.0,
      "step": 825
    },
    {
      "entropy": 7.084704065322876,
      "epoch": 0.7133648474430597,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004145,
      "loss": 6.9896,
      "mean_token_accuracy": 0.11336685940623284,
      "num_tokens": 1509698.0,
      "step": 830
    },
    {
      "entropy": 7.083193683624268,
      "epoch": 0.7176622260421143,
      "grad_norm": 1.125,
      "learning_rate": 0.000417,
      "loss": 6.9501,
      "mean_token_accuracy": 0.1161960445344448,
      "num_tokens": 1518224.0,
      "step": 835
    },
    {
      "entropy": 7.068277740478516,
      "epoch": 0.7219596046411689,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004195,
      "loss": 6.9509,
      "mean_token_accuracy": 0.11460287198424339,
      "num_tokens": 1527163.0,
      "step": 840
    },
    {
      "entropy": 7.035658502578736,
      "epoch": 0.7262569832402235,
      "grad_norm": 1.1875,
      "learning_rate": 0.000422,
      "loss": 6.9773,
      "mean_token_accuracy": 0.11205969154834747,
      "num_tokens": 1536980.0,
      "step": 845
    },
    {
      "entropy": 7.014459228515625,
      "epoch": 0.730554361839278,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 6.933,
      "mean_token_accuracy": 0.1107229508459568,
      "num_tokens": 1546388.0,
      "step": 850
    },
    {
      "entropy": 7.04270453453064,
      "epoch": 0.7348517404383326,
      "grad_norm": 1.171875,
      "learning_rate": 0.000427,
      "loss": 6.9719,
      "mean_token_accuracy": 0.1091302290558815,
      "num_tokens": 1555618.0,
      "step": 855
    },
    {
      "entropy": 7.163297176361084,
      "epoch": 0.7391491190373872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004295,
      "loss": 7.0471,
      "mean_token_accuracy": 0.11037192791700363,
      "num_tokens": 1565398.0,
      "step": 860
    },
    {
      "entropy": 7.015673398971558,
      "epoch": 0.7434464976364418,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432,
      "loss": 6.9003,
      "mean_token_accuracy": 0.11521378830075264,
      "num_tokens": 1574052.0,
      "step": 865
    },
    {
      "entropy": 6.948957300186157,
      "epoch": 0.7477438762354963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004345,
      "loss": 6.8385,
      "mean_token_accuracy": 0.12051832377910614,
      "num_tokens": 1582930.0,
      "step": 870
    },
    {
      "entropy": 6.9927011013031,
      "epoch": 0.752041254834551,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437,
      "loss": 6.9591,
      "mean_token_accuracy": 0.11156816780567169,
      "num_tokens": 1592007.0,
      "step": 875
    },
    {
      "entropy": 7.008735847473145,
      "epoch": 0.7563386334336055,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004395,
      "loss": 6.7845,
      "mean_token_accuracy": 0.12361353784799575,
      "num_tokens": 1600532.0,
      "step": 880
    },
    {
      "entropy": 6.962292051315307,
      "epoch": 0.7606360120326601,
      "grad_norm": 1.125,
      "learning_rate": 0.000442,
      "loss": 6.9289,
      "mean_token_accuracy": 0.1097280241549015,
      "num_tokens": 1610319.0,
      "step": 885
    },
    {
      "entropy": 6.939612913131714,
      "epoch": 0.7649333906317146,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004445,
      "loss": 6.7976,
      "mean_token_accuracy": 0.11956171989440918,
      "num_tokens": 1618161.0,
      "step": 890
    },
    {
      "entropy": 7.0276683330535885,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000447,
      "loss": 6.9023,
      "mean_token_accuracy": 0.11797274127602578,
      "num_tokens": 1627676.0,
      "step": 895
    },
    {
      "entropy": 6.978840589523315,
      "epoch": 0.7735281478298238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.8601,
      "mean_token_accuracy": 0.1139043152332306,
      "num_tokens": 1636693.0,
      "step": 900
    },
    {
      "entropy": 6.861077356338501,
      "epoch": 0.7778255264288784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.8644,
      "mean_token_accuracy": 0.11552541553974152,
      "num_tokens": 1645577.0,
      "step": 905
    },
    {
      "entropy": 6.998029899597168,
      "epoch": 0.7821229050279329,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.8572,
      "mean_token_accuracy": 0.1161652997136116,
      "num_tokens": 1654239.0,
      "step": 910
    },
    {
      "entropy": 7.036835432052612,
      "epoch": 0.7864202836269876,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.8756,
      "mean_token_accuracy": 0.11719652488827706,
      "num_tokens": 1663085.0,
      "step": 915
    },
    {
      "entropy": 6.875935173034668,
      "epoch": 0.7907176622260421,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.9058,
      "mean_token_accuracy": 0.11936888322234154,
      "num_tokens": 1671621.0,
      "step": 920
    },
    {
      "entropy": 7.0160260677337645,
      "epoch": 0.7950150408250967,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000462,
      "loss": 6.8876,
      "mean_token_accuracy": 0.11704642698168755,
      "num_tokens": 1680912.0,
      "step": 925
    },
    {
      "entropy": 6.802255773544312,
      "epoch": 0.7993124194241513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004645,
      "loss": 6.7984,
      "mean_token_accuracy": 0.1263578489422798,
      "num_tokens": 1689772.0,
      "step": 930
    },
    {
      "entropy": 7.00699782371521,
      "epoch": 0.8036097980232059,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000467,
      "loss": 6.9767,
      "mean_token_accuracy": 0.10714940577745438,
      "num_tokens": 1698985.0,
      "step": 935
    },
    {
      "entropy": 6.9496043682098385,
      "epoch": 0.8079071766222604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004695,
      "loss": 6.8334,
      "mean_token_accuracy": 0.11713856160640716,
      "num_tokens": 1707365.0,
      "step": 940
    },
    {
      "entropy": 6.864590072631836,
      "epoch": 0.812204555221315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000472,
      "loss": 6.8486,
      "mean_token_accuracy": 0.12177055031061172,
      "num_tokens": 1715858.0,
      "step": 945
    },
    {
      "entropy": 7.043800115585327,
      "epoch": 0.8165019338203696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004745,
      "loss": 6.8787,
      "mean_token_accuracy": 0.11329701766371728,
      "num_tokens": 1724721.0,
      "step": 950
    },
    {
      "entropy": 6.8629217624664305,
      "epoch": 0.8207993124194242,
      "grad_norm": 1.1875,
      "learning_rate": 0.000477,
      "loss": 6.8973,
      "mean_token_accuracy": 0.11501588821411132,
      "num_tokens": 1733700.0,
      "step": 955
    },
    {
      "entropy": 6.911140823364258,
      "epoch": 0.8250966910184787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004795,
      "loss": 6.8497,
      "mean_token_accuracy": 0.11782671734690667,
      "num_tokens": 1742167.0,
      "step": 960
    },
    {
      "entropy": 7.022890186309814,
      "epoch": 0.8293940696175333,
      "grad_norm": 1.125,
      "learning_rate": 0.000482,
      "loss": 6.8634,
      "mean_token_accuracy": 0.11890430226922036,
      "num_tokens": 1752087.0,
      "step": 965
    },
    {
      "entropy": 6.841674852371216,
      "epoch": 0.8336914482165879,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004845,
      "loss": 6.8119,
      "mean_token_accuracy": 0.12273848205804824,
      "num_tokens": 1760749.0,
      "step": 970
    },
    {
      "entropy": 6.708367443084716,
      "epoch": 0.8379888268156425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000487,
      "loss": 6.8394,
      "mean_token_accuracy": 0.118587414175272,
      "num_tokens": 1770923.0,
      "step": 975
    },
    {
      "entropy": 6.947722911834717,
      "epoch": 0.842286205414697,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004895,
      "loss": 6.8257,
      "mean_token_accuracy": 0.11907628253102302,
      "num_tokens": 1779233.0,
      "step": 980
    },
    {
      "entropy": 6.948696708679199,
      "epoch": 0.8465835840137517,
      "grad_norm": 1.375,
      "learning_rate": 0.000492,
      "loss": 6.8234,
      "mean_token_accuracy": 0.11929655894637108,
      "num_tokens": 1787593.0,
      "step": 985
    },
    {
      "entropy": 6.848589468002319,
      "epoch": 0.8508809626128062,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004945,
      "loss": 6.8683,
      "mean_token_accuracy": 0.11564295887947082,
      "num_tokens": 1796655.0,
      "step": 990
    },
    {
      "entropy": 6.848339986801148,
      "epoch": 0.8551783412118608,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000497,
      "loss": 6.7993,
      "mean_token_accuracy": 0.1252160184085369,
      "num_tokens": 1805790.0,
      "step": 995
    },
    {
      "entropy": 6.861878204345703,
      "epoch": 0.8594757198109153,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995,
      "loss": 6.7821,
      "mean_token_accuracy": 0.11595688387751579,
      "num_tokens": 1813776.0,
      "step": 1000
    },
    {
      "epoch": 0.8594757198109153,
      "eval_entropy": 6.850721193863465,
      "eval_loss": 6.848105430603027,
      "eval_mean_token_accuracy": 0.12149045062628952,
      "eval_num_tokens": 1813776.0,
      "eval_runtime": 2.8306,
      "eval_samples_per_second": 1253.81,
      "eval_steps_per_second": 156.859,
      "step": 1000
    },
    {
      "entropy": 6.787899780273437,
      "epoch": 0.86377309840997,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999998427807679,
      "loss": 6.7924,
      "mean_token_accuracy": 0.11906375139951705,
      "num_tokens": 1823034.0,
      "step": 1005
    },
    {
      "entropy": 6.925178575515747,
      "epoch": 0.8680704770090245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999992040780138,
      "loss": 6.8714,
      "mean_token_accuracy": 0.115791717171669,
      "num_tokens": 1832320.0,
      "step": 1010
    },
    {
      "entropy": 6.781938219070435,
      "epoch": 0.8723678556080791,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999980740669294,
      "loss": 6.7573,
      "mean_token_accuracy": 0.12310137376189231,
      "num_tokens": 1841165.0,
      "step": 1015
    },
    {
      "entropy": 6.863918399810791,
      "epoch": 0.8766652342071336,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999964527499823,
      "loss": 6.8264,
      "mean_token_accuracy": 0.11974046677350998,
      "num_tokens": 1850440.0,
      "step": 1020
    },
    {
      "entropy": 6.80152702331543,
      "epoch": 0.8809626128061883,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004999943401307127,
      "loss": 6.8336,
      "mean_token_accuracy": 0.11709848865866661,
      "num_tokens": 1859304.0,
      "step": 1025
    },
    {
      "entropy": 6.884733772277832,
      "epoch": 0.8852599914052428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999917362137337,
      "loss": 6.7027,
      "mean_token_accuracy": 0.12448290213942528,
      "num_tokens": 1867560.0,
      "step": 1030
    },
    {
      "entropy": 6.801468753814698,
      "epoch": 0.8895573700042974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999886410047312,
      "loss": 6.7809,
      "mean_token_accuracy": 0.12414068430662155,
      "num_tokens": 1876742.0,
      "step": 1035
    },
    {
      "entropy": 6.889640283584595,
      "epoch": 0.8938547486033519,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999850545104638,
      "loss": 6.7848,
      "mean_token_accuracy": 0.12225122228264809,
      "num_tokens": 1885738.0,
      "step": 1040
    },
    {
      "entropy": 6.826289463043213,
      "epoch": 0.8981521272024066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999809767387633,
      "loss": 6.8167,
      "mean_token_accuracy": 0.11297583729028701,
      "num_tokens": 1894688.0,
      "step": 1045
    },
    {
      "entropy": 6.887860727310181,
      "epoch": 0.9024495058014611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999764076985337,
      "loss": 6.8767,
      "mean_token_accuracy": 0.12184796631336212,
      "num_tokens": 1903809.0,
      "step": 1050
    },
    {
      "entropy": 6.873423671722412,
      "epoch": 0.9067468844005157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999713473997519,
      "loss": 6.834,
      "mean_token_accuracy": 0.1166172593832016,
      "num_tokens": 1913558.0,
      "step": 1055
    },
    {
      "entropy": 6.811271381378174,
      "epoch": 0.9110442629995703,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999657958534677,
      "loss": 6.8109,
      "mean_token_accuracy": 0.11310545653104782,
      "num_tokens": 1923324.0,
      "step": 1060
    },
    {
      "entropy": 6.750212335586548,
      "epoch": 0.9153416415986249,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999597530718034,
      "loss": 6.7839,
      "mean_token_accuracy": 0.120677300542593,
      "num_tokens": 1932849.0,
      "step": 1065
    },
    {
      "entropy": 6.9223932266235355,
      "epoch": 0.9196390201976794,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499953219067954,
      "loss": 6.9007,
      "mean_token_accuracy": 0.11593888625502587,
      "num_tokens": 1942678.0,
      "step": 1070
    },
    {
      "entropy": 6.778281164169312,
      "epoch": 0.923936398796734,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999461938561873,
      "loss": 6.7536,
      "mean_token_accuracy": 0.11781821995973588,
      "num_tokens": 1951670.0,
      "step": 1075
    },
    {
      "entropy": 6.896324634552002,
      "epoch": 0.9282337773957886,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999386774518432,
      "loss": 6.8645,
      "mean_token_accuracy": 0.12120396941900254,
      "num_tokens": 1961060.0,
      "step": 1080
    },
    {
      "entropy": 6.905940532684326,
      "epoch": 0.9325311559948432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999306698713349,
      "loss": 6.8059,
      "mean_token_accuracy": 0.11919425427913666,
      "num_tokens": 1970790.0,
      "step": 1085
    },
    {
      "entropy": 6.860645055770874,
      "epoch": 0.9368285345938977,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999221711321477,
      "loss": 6.8232,
      "mean_token_accuracy": 0.11798580288887024,
      "num_tokens": 1980016.0,
      "step": 1090
    },
    {
      "entropy": 6.765835952758789,
      "epoch": 0.9411259131929522,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999131812528393,
      "loss": 6.7942,
      "mean_token_accuracy": 0.12255456075072288,
      "num_tokens": 1989542.0,
      "step": 1095
    },
    {
      "entropy": 6.922926950454712,
      "epoch": 0.9454232917920069,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049990370025304,
      "loss": 6.7695,
      "mean_token_accuracy": 0.1269351787865162,
      "num_tokens": 1999212.0,
      "step": 1100
    },
    {
      "entropy": 6.666564416885376,
      "epoch": 0.9497206703910615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998937281534526,
      "loss": 6.7551,
      "mean_token_accuracy": 0.12631533294916153,
      "num_tokens": 2008378.0,
      "step": 1105
    },
    {
      "entropy": 6.820988655090332,
      "epoch": 0.954018048990116,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004998832649758521,
      "loss": 6.8346,
      "mean_token_accuracy": 0.11563185527920723,
      "num_tokens": 2018985.0,
      "step": 1110
    },
    {
      "entropy": 6.789529275894165,
      "epoch": 0.9583154275891707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998723107430862,
      "loss": 6.7218,
      "mean_token_accuracy": 0.12006904929876328,
      "num_tokens": 2027274.0,
      "step": 1115
    },
    {
      "entropy": 6.760359621047973,
      "epoch": 0.9626128061882252,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998608654790741,
      "loss": 6.8206,
      "mean_token_accuracy": 0.12053585797548294,
      "num_tokens": 2036892.0,
      "step": 1120
    },
    {
      "entropy": 6.926661109924316,
      "epoch": 0.9669101847872797,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499848929208808,
      "loss": 6.7966,
      "mean_token_accuracy": 0.11386888101696968,
      "num_tokens": 2046042.0,
      "step": 1125
    },
    {
      "entropy": 6.7287068367004395,
      "epoch": 0.9712075633863343,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998365019583519,
      "loss": 6.7663,
      "mean_token_accuracy": 0.11801247894763947,
      "num_tokens": 2054576.0,
      "step": 1130
    },
    {
      "entropy": 6.731119585037232,
      "epoch": 0.975504941985389,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998235837548417,
      "loss": 6.7113,
      "mean_token_accuracy": 0.12452540397644044,
      "num_tokens": 2063552.0,
      "step": 1135
    },
    {
      "entropy": 6.721849632263184,
      "epoch": 0.9798023205844435,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499810174626486,
      "loss": 6.7168,
      "mean_token_accuracy": 0.12163661047816277,
      "num_tokens": 2072965.0,
      "step": 1140
    },
    {
      "entropy": 6.808212089538574,
      "epoch": 0.984099699183498,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004997962746025646,
      "loss": 6.7754,
      "mean_token_accuracy": 0.12033762335777283,
      "num_tokens": 2082461.0,
      "step": 1145
    },
    {
      "entropy": 6.747566175460816,
      "epoch": 0.9883970777825526,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997818837134298,
      "loss": 6.7715,
      "mean_token_accuracy": 0.12163460701704025,
      "num_tokens": 2091022.0,
      "step": 1150
    },
    {
      "entropy": 6.744241046905517,
      "epoch": 0.9926944563816072,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997670019905057,
      "loss": 6.6552,
      "mean_token_accuracy": 0.13288048580288886,
      "num_tokens": 2099896.0,
      "step": 1155
    },
    {
      "entropy": 6.769083023071289,
      "epoch": 0.9969918349806618,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997516294662876,
      "loss": 6.7577,
      "mean_token_accuracy": 0.12401864305138588,
      "num_tokens": 2109122.0,
      "step": 1160
    },
    {
      "entropy": 6.859591378106011,
      "epoch": 1.0008594757198108,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997357661743433,
      "loss": 6.7057,
      "mean_token_accuracy": 0.12527080294158724,
      "num_tokens": 2117879.0,
      "step": 1165
    },
    {
      "entropy": 6.636478424072266,
      "epoch": 1.0051568543188656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997194121493118,
      "loss": 6.4751,
      "mean_token_accuracy": 0.13114811182022096,
      "num_tokens": 2127552.0,
      "step": 1170
    },
    {
      "entropy": 6.738952016830444,
      "epoch": 1.0094542329179201,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997025674269037,
      "loss": 6.4838,
      "mean_token_accuracy": 0.12866360768675805,
      "num_tokens": 2136648.0,
      "step": 1175
    },
    {
      "entropy": 6.67201509475708,
      "epoch": 1.0137516115169747,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996852320439013,
      "loss": 6.5263,
      "mean_token_accuracy": 0.13279551938176154,
      "num_tokens": 2144852.0,
      "step": 1180
    },
    {
      "entropy": 6.675800752639771,
      "epoch": 1.0180489901160292,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996674060381578,
      "loss": 6.5531,
      "mean_token_accuracy": 0.12884859815239907,
      "num_tokens": 2154567.0,
      "step": 1185
    },
    {
      "entropy": 6.688924407958984,
      "epoch": 1.0223463687150838,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996490894485985,
      "loss": 6.5359,
      "mean_token_accuracy": 0.12873365879058837,
      "num_tokens": 2164260.0,
      "step": 1190
    },
    {
      "entropy": 6.644716072082519,
      "epoch": 1.0266437473141383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996302823152193,
      "loss": 6.5181,
      "mean_token_accuracy": 0.12822481170296668,
      "num_tokens": 2173247.0,
      "step": 1195
    },
    {
      "entropy": 6.574692296981811,
      "epoch": 1.0309411259131929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996109846790873,
      "loss": 6.5603,
      "mean_token_accuracy": 0.12655097916722297,
      "num_tokens": 2181595.0,
      "step": 1200
    },
    {
      "entropy": 6.753303909301758,
      "epoch": 1.0352385045122476,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995911965823412,
      "loss": 6.4769,
      "mean_token_accuracy": 0.12973184511065483,
      "num_tokens": 2190608.0,
      "step": 1205
    },
    {
      "entropy": 6.538211107254028,
      "epoch": 1.0395358831113022,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995709180681899,
      "loss": 6.4797,
      "mean_token_accuracy": 0.14387280195951463,
      "num_tokens": 2199131.0,
      "step": 1210
    },
    {
      "entropy": 6.674481058120728,
      "epoch": 1.0438332617103567,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499550149180914,
      "loss": 6.6247,
      "mean_token_accuracy": 0.12701441198587418,
      "num_tokens": 2208365.0,
      "step": 1215
    },
    {
      "entropy": 6.658597755432129,
      "epoch": 1.0481306403094113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995288899658641,
      "loss": 6.4953,
      "mean_token_accuracy": 0.1288551889359951,
      "num_tokens": 2216851.0,
      "step": 1220
    },
    {
      "entropy": 6.630938100814819,
      "epoch": 1.0524280189084658,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995071404694619,
      "loss": 6.5736,
      "mean_token_accuracy": 0.12576404958963394,
      "num_tokens": 2225309.0,
      "step": 1225
    },
    {
      "entropy": 6.695171594619751,
      "epoch": 1.0567253975075204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994849007391996,
      "loss": 6.5228,
      "mean_token_accuracy": 0.1320165403187275,
      "num_tokens": 2235036.0,
      "step": 1230
    },
    {
      "entropy": 6.625878429412841,
      "epoch": 1.061022776106575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994621708236401,
      "loss": 6.5198,
      "mean_token_accuracy": 0.12893396690487863,
      "num_tokens": 2244038.0,
      "step": 1235
    },
    {
      "entropy": 6.645451402664184,
      "epoch": 1.0653201547056295,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499438950772416,
      "loss": 6.4709,
      "mean_token_accuracy": 0.13499100208282472,
      "num_tokens": 2252828.0,
      "step": 1240
    },
    {
      "entropy": 6.610739994049072,
      "epoch": 1.0696175333046842,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994152406362311,
      "loss": 6.4523,
      "mean_token_accuracy": 0.13818272650241853,
      "num_tokens": 2261709.0,
      "step": 1245
    },
    {
      "entropy": 6.59524974822998,
      "epoch": 1.0739149119037388,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993910404668586,
      "loss": 6.4923,
      "mean_token_accuracy": 0.13403210416436195,
      "num_tokens": 2269801.0,
      "step": 1250
    },
    {
      "entropy": 6.591278791427612,
      "epoch": 1.0782122905027933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499366350317142,
      "loss": 6.4461,
      "mean_token_accuracy": 0.1370372988283634,
      "num_tokens": 2280587.0,
      "step": 1255
    },
    {
      "entropy": 6.460198545455933,
      "epoch": 1.0825096691018479,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993411702409948,
      "loss": 6.4397,
      "mean_token_accuracy": 0.1323759824037552,
      "num_tokens": 2289783.0,
      "step": 1260
    },
    {
      "entropy": 6.588274049758911,
      "epoch": 1.0868070477009024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993155002934002,
      "loss": 6.5313,
      "mean_token_accuracy": 0.12670424431562424,
      "num_tokens": 2298785.0,
      "step": 1265
    },
    {
      "entropy": 6.64993691444397,
      "epoch": 1.091104426299957,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992893405304111,
      "loss": 6.6208,
      "mean_token_accuracy": 0.13185966834425927,
      "num_tokens": 2308400.0,
      "step": 1270
    },
    {
      "entropy": 6.641791296005249,
      "epoch": 1.0954018048990115,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049926269100915,
      "loss": 6.4313,
      "mean_token_accuracy": 0.13738049939274788,
      "num_tokens": 2317440.0,
      "step": 1275
    },
    {
      "entropy": 6.503924989700318,
      "epoch": 1.0996991834980663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992355517878087,
      "loss": 6.4783,
      "mean_token_accuracy": 0.131829671561718,
      "num_tokens": 2326730.0,
      "step": 1280
    },
    {
      "entropy": 6.57679181098938,
      "epoch": 1.1039965620971208,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992079229256484,
      "loss": 6.4692,
      "mean_token_accuracy": 0.13649967312812805,
      "num_tokens": 2336126.0,
      "step": 1285
    },
    {
      "entropy": 6.6197991371154785,
      "epoch": 1.1082939406961754,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991798044829996,
      "loss": 6.4954,
      "mean_token_accuracy": 0.13550228029489517,
      "num_tokens": 2345144.0,
      "step": 1290
    },
    {
      "entropy": 6.5108154773712155,
      "epoch": 1.11259131929523,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991511965212618,
      "loss": 6.444,
      "mean_token_accuracy": 0.13719593361020088,
      "num_tokens": 2353851.0,
      "step": 1295
    },
    {
      "entropy": 6.5021027565002445,
      "epoch": 1.1168886978942845,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991220991029032,
      "loss": 6.4387,
      "mean_token_accuracy": 0.13497617840766907,
      "num_tokens": 2362316.0,
      "step": 1300
    },
    {
      "entropy": 6.660447835922241,
      "epoch": 1.121186076493339,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499092512291461,
      "loss": 6.5457,
      "mean_token_accuracy": 0.1298186369240284,
      "num_tokens": 2373350.0,
      "step": 1305
    },
    {
      "entropy": 6.587765121459961,
      "epoch": 1.1254834550923936,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499062436151541,
      "loss": 6.4847,
      "mean_token_accuracy": 0.13602642267942427,
      "num_tokens": 2381753.0,
      "step": 1310
    },
    {
      "entropy": 6.5865904808044435,
      "epoch": 1.129780833691448,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990318707488173,
      "loss": 6.5174,
      "mean_token_accuracy": 0.1304781563580036,
      "num_tokens": 2390093.0,
      "step": 1315
    },
    {
      "entropy": 6.707785511016846,
      "epoch": 1.1340782122905029,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990008161500327,
      "loss": 6.5307,
      "mean_token_accuracy": 0.13693483620882035,
      "num_tokens": 2399103.0,
      "step": 1320
    },
    {
      "entropy": 6.539694452285767,
      "epoch": 1.1383755908895574,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498969272422998,
      "loss": 6.4963,
      "mean_token_accuracy": 0.13442791253328323,
      "num_tokens": 2407530.0,
      "step": 1325
    },
    {
      "entropy": 6.5834043502807615,
      "epoch": 1.142672969488612,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989372396365921,
      "loss": 6.4721,
      "mean_token_accuracy": 0.13653133809566498,
      "num_tokens": 2416264.0,
      "step": 1330
    },
    {
      "entropy": 6.560483932495117,
      "epoch": 1.1469703480876665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989047178607618,
      "loss": 6.5643,
      "mean_token_accuracy": 0.1255203329026699,
      "num_tokens": 2425256.0,
      "step": 1335
    },
    {
      "entropy": 6.543246126174926,
      "epoch": 1.151267726686721,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988717071665215,
      "loss": 6.5303,
      "mean_token_accuracy": 0.12744321376085282,
      "num_tokens": 2434343.0,
      "step": 1340
    },
    {
      "entropy": 6.625397109985352,
      "epoch": 1.1555651052857756,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004988382076259537,
      "loss": 6.439,
      "mean_token_accuracy": 0.13895059525966644,
      "num_tokens": 2442864.0,
      "step": 1345
    },
    {
      "entropy": 6.505967378616333,
      "epoch": 1.1598624838848304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988042193122077,
      "loss": 6.482,
      "mean_token_accuracy": 0.1311720348894596,
      "num_tokens": 2451701.0,
      "step": 1350
    },
    {
      "entropy": 6.503494215011597,
      "epoch": 1.164159862483885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987697422995005,
      "loss": 6.4696,
      "mean_token_accuracy": 0.13065513744950294,
      "num_tokens": 2460943.0,
      "step": 1355
    },
    {
      "entropy": 6.595712518692016,
      "epoch": 1.1684572410829395,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987347766631161,
      "loss": 6.4954,
      "mean_token_accuracy": 0.13297019228339196,
      "num_tokens": 2470294.0,
      "step": 1360
    },
    {
      "entropy": 6.517033290863037,
      "epoch": 1.172754619681994,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004986993224794055,
      "loss": 6.5629,
      "mean_token_accuracy": 0.13079115897417068,
      "num_tokens": 2479530.0,
      "step": 1365
    },
    {
      "entropy": 6.589979982376098,
      "epoch": 1.1770519982810486,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986633798257865,
      "loss": 6.3268,
      "mean_token_accuracy": 0.13975912407040597,
      "num_tokens": 2488511.0,
      "step": 1370
    },
    {
      "entropy": 6.550735712051392,
      "epoch": 1.181349376880103,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986269487807434,
      "loss": 6.5284,
      "mean_token_accuracy": 0.1279216580092907,
      "num_tokens": 2497868.0,
      "step": 1375
    },
    {
      "entropy": 6.562166547775268,
      "epoch": 1.1856467554791577,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498590029423827,
      "loss": 6.4477,
      "mean_token_accuracy": 0.13719073608517646,
      "num_tokens": 2506873.0,
      "step": 1380
    },
    {
      "entropy": 6.647982025146485,
      "epoch": 1.1899441340782122,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985526218356546,
      "loss": 6.5799,
      "mean_token_accuracy": 0.13164269253611566,
      "num_tokens": 2517317.0,
      "step": 1385
    },
    {
      "entropy": 6.559144067764282,
      "epoch": 1.1942415126772667,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985147260979093,
      "loss": 6.4983,
      "mean_token_accuracy": 0.13328756764531136,
      "num_tokens": 2527348.0,
      "step": 1390
    },
    {
      "entropy": 6.5675629615783695,
      "epoch": 1.1985388912763215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004984763422933402,
      "loss": 6.4813,
      "mean_token_accuracy": 0.13246413692831993,
      "num_tokens": 2536277.0,
      "step": 1395
    },
    {
      "entropy": 6.5679216384887695,
      "epoch": 1.202836269875376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984374705057623,
      "loss": 6.5581,
      "mean_token_accuracy": 0.1256929397583008,
      "num_tokens": 2545638.0,
      "step": 1400
    },
    {
      "entropy": 6.445409727096558,
      "epoch": 1.2071336484744306,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983981108200561,
      "loss": 6.3901,
      "mean_token_accuracy": 0.13980056941509247,
      "num_tokens": 2555694.0,
      "step": 1405
    },
    {
      "entropy": 6.59971604347229,
      "epoch": 1.2114310270734852,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983582633221672,
      "loss": 6.4436,
      "mean_token_accuracy": 0.13092342391610146,
      "num_tokens": 2564344.0,
      "step": 1410
    },
    {
      "entropy": 6.468144750595092,
      "epoch": 1.2157284056725397,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983179280991068,
      "loss": 6.4329,
      "mean_token_accuracy": 0.13643699064850806,
      "num_tokens": 2573726.0,
      "step": 1415
    },
    {
      "entropy": 6.531007194519043,
      "epoch": 1.2200257842715942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982771052389508,
      "loss": 6.4943,
      "mean_token_accuracy": 0.13096859753131868,
      "num_tokens": 2582845.0,
      "step": 1420
    },
    {
      "entropy": 6.531826734542847,
      "epoch": 1.224323162870649,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982357948308401,
      "loss": 6.3872,
      "mean_token_accuracy": 0.1383131504058838,
      "num_tokens": 2591258.0,
      "step": 1425
    },
    {
      "entropy": 6.575545120239258,
      "epoch": 1.2286205414697036,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004981939969649799,
      "loss": 6.4782,
      "mean_token_accuracy": 0.13248563408851624,
      "num_tokens": 2600099.0,
      "step": 1430
    },
    {
      "entropy": 6.531699705123901,
      "epoch": 1.232917920068758,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981517117326404,
      "loss": 6.4466,
      "mean_token_accuracy": 0.1367930218577385,
      "num_tokens": 2608845.0,
      "step": 1435
    },
    {
      "entropy": 6.4251275062561035,
      "epoch": 1.2372152986678127,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981089392261553,
      "loss": 6.3915,
      "mean_token_accuracy": 0.14260655716061593,
      "num_tokens": 2618723.0,
      "step": 1440
    },
    {
      "entropy": 6.515915298461914,
      "epoch": 1.2415126772668672,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498065679538923,
      "loss": 6.4512,
      "mean_token_accuracy": 0.13224708959460257,
      "num_tokens": 2628542.0,
      "step": 1445
    },
    {
      "entropy": 6.4922788619995115,
      "epoch": 1.2458100558659218,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980219327654049,
      "loss": 6.4214,
      "mean_token_accuracy": 0.14079917669296266,
      "num_tokens": 2637395.0,
      "step": 1450
    },
    {
      "entropy": 6.43779935836792,
      "epoch": 1.2501074344649763,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497977699001127,
      "loss": 6.2848,
      "mean_token_accuracy": 0.14780827164649962,
      "num_tokens": 2645545.0,
      "step": 1455
    },
    {
      "entropy": 6.449300241470337,
      "epoch": 1.2544048130640308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979329783426778,
      "loss": 6.4446,
      "mean_token_accuracy": 0.1389339432120323,
      "num_tokens": 2654000.0,
      "step": 1460
    },
    {
      "entropy": 6.54451584815979,
      "epoch": 1.2587021916630854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978877708877094,
      "loss": 6.4506,
      "mean_token_accuracy": 0.13394884243607522,
      "num_tokens": 2664017.0,
      "step": 1465
    },
    {
      "entropy": 6.453659868240356,
      "epoch": 1.2629995702621402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978420767349368,
      "loss": 6.4412,
      "mean_token_accuracy": 0.13818295150995255,
      "num_tokens": 2673250.0,
      "step": 1470
    },
    {
      "entropy": 6.53358006477356,
      "epoch": 1.2672969488611947,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977958959841379,
      "loss": 6.4796,
      "mean_token_accuracy": 0.13405440375208855,
      "num_tokens": 2682190.0,
      "step": 1475
    },
    {
      "entropy": 6.550917911529541,
      "epoch": 1.2715943274602493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000497749228736153,
      "loss": 6.4152,
      "mean_token_accuracy": 0.13004215285182,
      "num_tokens": 2691425.0,
      "step": 1480
    },
    {
      "entropy": 6.5540321350097654,
      "epoch": 1.2758917060593038,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004977020750928845,
      "loss": 6.5022,
      "mean_token_accuracy": 0.1328156165778637,
      "num_tokens": 2701123.0,
      "step": 1485
    },
    {
      "entropy": 6.401340389251709,
      "epoch": 1.2801890846583583,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976544351572973,
      "loss": 6.5081,
      "mean_token_accuracy": 0.1309308812022209,
      "num_tokens": 2710829.0,
      "step": 1490
    },
    {
      "entropy": 6.600834655761719,
      "epoch": 1.2844864632574131,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004976063090334179,
      "loss": 6.4494,
      "mean_token_accuracy": 0.13603252917528152,
      "num_tokens": 2720313.0,
      "step": 1495
    },
    {
      "entropy": 6.446039533615112,
      "epoch": 1.2887838418564677,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004975576968263346,
      "loss": 6.3644,
      "mean_token_accuracy": 0.13654347658157348,
      "num_tokens": 2729734.0,
      "step": 1500
    },
    {
      "epoch": 1.2887838418564677,
      "eval_entropy": 6.42241061270774,
      "eval_loss": 6.562856674194336,
      "eval_mean_token_accuracy": 0.13629841859880332,
      "eval_num_tokens": 2729734.0,
      "eval_runtime": 2.6412,
      "eval_samples_per_second": 1343.717,
      "eval_steps_per_second": 168.107,
      "step": 1500
    },
    {
      "entropy": 6.496413612365723,
      "epoch": 1.2930812204555222,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497508598642197,
      "loss": 6.4498,
      "mean_token_accuracy": 0.13591181635856628,
      "num_tokens": 2739001.0,
      "step": 1505
    },
    {
      "entropy": 6.4927387714385985,
      "epoch": 1.2973785990545768,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497459014588216,
      "loss": 6.3461,
      "mean_token_accuracy": 0.14019913077354432,
      "num_tokens": 2748460.0,
      "step": 1510
    },
    {
      "entropy": 6.442285966873169,
      "epoch": 1.3016759776536313,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497408944772663,
      "loss": 6.4495,
      "mean_token_accuracy": 0.13356760740280152,
      "num_tokens": 2757993.0,
      "step": 1515
    },
    {
      "entropy": 6.442544317245483,
      "epoch": 1.3059733562526858,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973583893048707,
      "loss": 6.4679,
      "mean_token_accuracy": 0.1391291558742523,
      "num_tokens": 2767380.0,
      "step": 1520
    },
    {
      "entropy": 6.537827825546264,
      "epoch": 1.3102707348517404,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973073482952321,
      "loss": 6.3971,
      "mean_token_accuracy": 0.1322616696357727,
      "num_tokens": 2776207.0,
      "step": 1525
    },
    {
      "entropy": 6.452885007858276,
      "epoch": 1.314568113450795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972558218552004,
      "loss": 6.4398,
      "mean_token_accuracy": 0.13432919085025788,
      "num_tokens": 2785857.0,
      "step": 1530
    },
    {
      "entropy": 6.486521863937378,
      "epoch": 1.3188654920498495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972038100972885,
      "loss": 6.5342,
      "mean_token_accuracy": 0.1303657077252865,
      "num_tokens": 2795500.0,
      "step": 1535
    },
    {
      "entropy": 6.549186372756958,
      "epoch": 1.323162870648904,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004971513131350697,
      "loss": 6.4088,
      "mean_token_accuracy": 0.13469692692160606,
      "num_tokens": 2804188.0,
      "step": 1540
    },
    {
      "entropy": 6.489724349975586,
      "epoch": 1.3274602492479588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004970983310831759,
      "loss": 6.424,
      "mean_token_accuracy": 0.12920740991830826,
      "num_tokens": 2814255.0,
      "step": 1545
    },
    {
      "entropy": 6.468247175216675,
      "epoch": 1.3317576278470133,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004970448640572989,
      "loss": 6.3908,
      "mean_token_accuracy": 0.13704784587025642,
      "num_tokens": 2823059.0,
      "step": 1550
    },
    {
      "entropy": 6.443971824645996,
      "epoch": 1.336055006446068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004969909121741895,
      "loss": 6.3212,
      "mean_token_accuracy": 0.14167552590370178,
      "num_tokens": 2831729.0,
      "step": 1555
    },
    {
      "entropy": 6.461103248596191,
      "epoch": 1.3403523850451224,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969364755516569,
      "loss": 6.3786,
      "mean_token_accuracy": 0.1437497019767761,
      "num_tokens": 2839597.0,
      "step": 1560
    },
    {
      "entropy": 6.413368654251099,
      "epoch": 1.344649763644177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968815543085689,
      "loss": 6.2597,
      "mean_token_accuracy": 0.1471225917339325,
      "num_tokens": 2847964.0,
      "step": 1565
    },
    {
      "entropy": 6.442201375961304,
      "epoch": 1.3489471422432318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968261485648516,
      "loss": 6.4682,
      "mean_token_accuracy": 0.13341710045933725,
      "num_tokens": 2857712.0,
      "step": 1570
    },
    {
      "entropy": 6.399351978302002,
      "epoch": 1.3532445208422863,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496770258441489,
      "loss": 6.4641,
      "mean_token_accuracy": 0.13078686520457267,
      "num_tokens": 2866724.0,
      "step": 1575
    },
    {
      "entropy": 6.548488187789917,
      "epoch": 1.3575418994413408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967138840605228,
      "loss": 6.3578,
      "mean_token_accuracy": 0.14059459790587425,
      "num_tokens": 2875318.0,
      "step": 1580
    },
    {
      "entropy": 6.420933103561401,
      "epoch": 1.3618392780403954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000496657025545052,
      "loss": 6.3964,
      "mean_token_accuracy": 0.13322596848011017,
      "num_tokens": 2883794.0,
      "step": 1585
    },
    {
      "entropy": 6.4831382751464846,
      "epoch": 1.36613665663945,
      "grad_norm": 1.171875,
      "learning_rate": 0.000496599683019233,
      "loss": 6.4175,
      "mean_token_accuracy": 0.1317307434976101,
      "num_tokens": 2891996.0,
      "step": 1590
    },
    {
      "entropy": 6.506607484817505,
      "epoch": 1.3704340352385045,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496541856608279,
      "loss": 6.5069,
      "mean_token_accuracy": 0.13520395755767822,
      "num_tokens": 2901718.0,
      "step": 1595
    },
    {
      "entropy": 6.488222885131836,
      "epoch": 1.374731413837559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964835464384595,
      "loss": 6.3369,
      "mean_token_accuracy": 0.1403390385210514,
      "num_tokens": 2910924.0,
      "step": 1600
    },
    {
      "entropy": 6.356355381011963,
      "epoch": 1.3790287924366136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000496424752637101,
      "loss": 6.4404,
      "mean_token_accuracy": 0.1325266629457474,
      "num_tokens": 2920075.0,
      "step": 1605
    },
    {
      "entropy": 6.4637175559997555,
      "epoch": 1.3833261710356681,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004963654753325853,
      "loss": 6.3958,
      "mean_token_accuracy": 0.14068760275840758,
      "num_tokens": 2928946.0,
      "step": 1610
    },
    {
      "entropy": 6.382746267318725,
      "epoch": 1.387623549634723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004963057146543505,
      "loss": 6.3788,
      "mean_token_accuracy": 0.14359928518533707,
      "num_tokens": 2936817.0,
      "step": 1615
    },
    {
      "entropy": 6.462194156646729,
      "epoch": 1.3919209282337774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049624547073289,
      "loss": 6.3627,
      "mean_token_accuracy": 0.1397081971168518,
      "num_tokens": 2946096.0,
      "step": 1620
    },
    {
      "entropy": 6.3689014434814455,
      "epoch": 1.396218306832832,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004961847436997526,
      "loss": 6.2999,
      "mean_token_accuracy": 0.147130286693573,
      "num_tokens": 2955632.0,
      "step": 1625
    },
    {
      "entropy": 6.4761738777160645,
      "epoch": 1.4005156854318865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961235336875416,
      "loss": 6.3289,
      "mean_token_accuracy": 0.14046488851308822,
      "num_tokens": 2963779.0,
      "step": 1630
    },
    {
      "entropy": 6.419768238067627,
      "epoch": 1.404813064030941,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004960618408299154,
      "loss": 6.3199,
      "mean_token_accuracy": 0.14862223267555236,
      "num_tokens": 2972778.0,
      "step": 1635
    },
    {
      "entropy": 6.360955905914307,
      "epoch": 1.4091104426299956,
      "grad_norm": 1.125,
      "learning_rate": 0.0004959996652615865,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1319722667336464,
      "num_tokens": 2982318.0,
      "step": 1640
    },
    {
      "entropy": 6.514253997802735,
      "epoch": 1.4134078212290504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959370071183216,
      "loss": 6.4039,
      "mean_token_accuracy": 0.14093454405665398,
      "num_tokens": 2991822.0,
      "step": 1645
    },
    {
      "entropy": 6.470624685287476,
      "epoch": 1.417705199828105,
      "grad_norm": 1.25,
      "learning_rate": 0.0004958738665369407,
      "loss": 6.4473,
      "mean_token_accuracy": 0.13844559118151664,
      "num_tokens": 3001203.0,
      "step": 1650
    },
    {
      "entropy": 6.412876272201538,
      "epoch": 1.4220025784271595,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004958102436553179,
      "loss": 6.3646,
      "mean_token_accuracy": 0.13862057998776436,
      "num_tokens": 3010445.0,
      "step": 1655
    },
    {
      "entropy": 6.440686845779419,
      "epoch": 1.426299957026214,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049574613861238,
      "loss": 6.3898,
      "mean_token_accuracy": 0.13541134893894197,
      "num_tokens": 3019919.0,
      "step": 1660
    },
    {
      "entropy": 6.509662914276123,
      "epoch": 1.4305973356252686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956815515481069,
      "loss": 6.5165,
      "mean_token_accuracy": 0.12789023146033288,
      "num_tokens": 3029469.0,
      "step": 1665
    },
    {
      "entropy": 6.461795949935913,
      "epoch": 1.4348947142243231,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004956164826035309,
      "loss": 6.3809,
      "mean_token_accuracy": 0.14309595823287963,
      "num_tokens": 3038627.0,
      "step": 1670
    },
    {
      "entropy": 6.41576566696167,
      "epoch": 1.4391920928233777,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955509319207363,
      "loss": 6.4312,
      "mean_token_accuracy": 0.13531435057520866,
      "num_tokens": 3048124.0,
      "step": 1675
    },
    {
      "entropy": 6.411845588684082,
      "epoch": 1.4434894714224322,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004954848996428601,
      "loss": 6.269,
      "mean_token_accuracy": 0.14824977070093154,
      "num_tokens": 3056394.0,
      "step": 1680
    },
    {
      "entropy": 6.383704996109008,
      "epoch": 1.4477868500214868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00049541838591409,
      "loss": 6.3767,
      "mean_token_accuracy": 0.1388603314757347,
      "num_tokens": 3066521.0,
      "step": 1685
    },
    {
      "entropy": 6.452395057678222,
      "epoch": 1.4520842286205415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004953513908796657,
      "loss": 6.4368,
      "mean_token_accuracy": 0.13928449675440788,
      "num_tokens": 3076571.0,
      "step": 1690
    },
    {
      "entropy": 6.397390127182007,
      "epoch": 1.456381607219596,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952839146858773,
      "loss": 6.4782,
      "mean_token_accuracy": 0.1314165897667408,
      "num_tokens": 3086086.0,
      "step": 1695
    },
    {
      "entropy": 6.3878926753997805,
      "epoch": 1.4606789858186506,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004952159574800658,
      "loss": 6.3599,
      "mean_token_accuracy": 0.14619865864515305,
      "num_tokens": 3095046.0,
      "step": 1700
    },
    {
      "entropy": 6.403419876098633,
      "epoch": 1.4649763644177052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004951475194106229,
      "loss": 6.3232,
      "mean_token_accuracy": 0.13754306733608246,
      "num_tokens": 3104193.0,
      "step": 1705
    },
    {
      "entropy": 6.475354766845703,
      "epoch": 1.4692737430167597,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950786006269898,
      "loss": 6.4103,
      "mean_token_accuracy": 0.1397714652121067,
      "num_tokens": 3112965.0,
      "step": 1710
    },
    {
      "entropy": 6.420578479766846,
      "epoch": 1.4735711216158143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004950092012796576,
      "loss": 6.4802,
      "mean_token_accuracy": 0.13525724932551383,
      "num_tokens": 3122883.0,
      "step": 1715
    },
    {
      "entropy": 6.446125411987305,
      "epoch": 1.477868500214869,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004949393215201666,
      "loss": 6.334,
      "mean_token_accuracy": 0.14860173985362052,
      "num_tokens": 3132182.0,
      "step": 1720
    },
    {
      "entropy": 6.438638639450073,
      "epoch": 1.4821658788139236,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948689615011065,
      "loss": 6.3881,
      "mean_token_accuracy": 0.13537878468632697,
      "num_tokens": 3141957.0,
      "step": 1725
    },
    {
      "entropy": 6.339945554733276,
      "epoch": 1.4864632574129781,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004947981213761154,
      "loss": 6.3201,
      "mean_token_accuracy": 0.1484260804951191,
      "num_tokens": 3151374.0,
      "step": 1730
    },
    {
      "entropy": 6.324282693862915,
      "epoch": 1.4907606360120327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947268012998797,
      "loss": 6.3382,
      "mean_token_accuracy": 0.13972473591566087,
      "num_tokens": 3160815.0,
      "step": 1735
    },
    {
      "entropy": 6.4526674270629885,
      "epoch": 1.4950580146110872,
      "grad_norm": 1.078125,
      "learning_rate": 0.000494655001428134,
      "loss": 6.345,
      "mean_token_accuracy": 0.14593861922621726,
      "num_tokens": 3169684.0,
      "step": 1740
    },
    {
      "entropy": 6.443645286560058,
      "epoch": 1.4993553932101418,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004945827219176604,
      "loss": 6.3344,
      "mean_token_accuracy": 0.13984118551015853,
      "num_tokens": 3178754.0,
      "step": 1745
    },
    {
      "entropy": 6.419276237487793,
      "epoch": 1.5036527718091963,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004945099629262888,
      "loss": 6.4109,
      "mean_token_accuracy": 0.1383548989892006,
      "num_tokens": 3187315.0,
      "step": 1750
    },
    {
      "entropy": 6.409943437576294,
      "epoch": 1.5079501504082509,
      "grad_norm": 1.125,
      "learning_rate": 0.0004944367246128954,
      "loss": 6.2452,
      "mean_token_accuracy": 0.1493537425994873,
      "num_tokens": 3195975.0,
      "step": 1755
    },
    {
      "entropy": 6.326399850845337,
      "epoch": 1.5122475290073054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943630071374036,
      "loss": 6.3539,
      "mean_token_accuracy": 0.14070346504449843,
      "num_tokens": 3204071.0,
      "step": 1760
    },
    {
      "entropy": 6.41026463508606,
      "epoch": 1.51654490760636,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942888106607828,
      "loss": 6.339,
      "mean_token_accuracy": 0.14475558772683145,
      "num_tokens": 3213001.0,
      "step": 1765
    },
    {
      "entropy": 6.4736912727355955,
      "epoch": 1.5208422862054147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942141353450486,
      "loss": 6.3736,
      "mean_token_accuracy": 0.13328379541635513,
      "num_tokens": 3222576.0,
      "step": 1770
    },
    {
      "entropy": 6.408621740341187,
      "epoch": 1.5251396648044693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004941389813532619,
      "loss": 6.3922,
      "mean_token_accuracy": 0.1397964984178543,
      "num_tokens": 3230474.0,
      "step": 1775
    },
    {
      "entropy": 6.462115287780762,
      "epoch": 1.5294370434035238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000494063348849529,
      "loss": 6.3624,
      "mean_token_accuracy": 0.14171589016914368,
      "num_tokens": 3238723.0,
      "step": 1780
    },
    {
      "entropy": 6.310955953598023,
      "epoch": 1.5337344220025786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004939872379990011,
      "loss": 6.3057,
      "mean_token_accuracy": 0.1414056584239006,
      "num_tokens": 3247844.0,
      "step": 1785
    },
    {
      "entropy": 6.499557447433472,
      "epoch": 1.5380318006016331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939106489678739,
      "loss": 6.3996,
      "mean_token_accuracy": 0.13961976021528244,
      "num_tokens": 3257228.0,
      "step": 1790
    },
    {
      "entropy": 6.428326511383057,
      "epoch": 1.5423291792006877,
      "grad_norm": 1.109375,
      "learning_rate": 0.000493833581923387,
      "loss": 6.3806,
      "mean_token_accuracy": 0.1350629784166813,
      "num_tokens": 3266655.0,
      "step": 1795
    },
    {
      "entropy": 6.40258674621582,
      "epoch": 1.5466265577997422,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004937560370338244,
      "loss": 6.3076,
      "mean_token_accuracy": 0.13736466988921164,
      "num_tokens": 3275213.0,
      "step": 1800
    },
    {
      "entropy": 6.4639317989349365,
      "epoch": 1.5509239363987968,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493678014468513,
      "loss": 6.4331,
      "mean_token_accuracy": 0.13448011130094528,
      "num_tokens": 3284447.0,
      "step": 1805
    },
    {
      "entropy": 6.456356191635132,
      "epoch": 1.5552213149978513,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004935995143978227,
      "loss": 6.3597,
      "mean_token_accuracy": 0.13500829488039018,
      "num_tokens": 3293568.0,
      "step": 1810
    },
    {
      "entropy": 6.347956132888794,
      "epoch": 1.5595186935969059,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004935205369931664,
      "loss": 6.2955,
      "mean_token_accuracy": 0.14433372840285302,
      "num_tokens": 3302055.0,
      "step": 1815
    },
    {
      "entropy": 6.415659618377686,
      "epoch": 1.5638160721959604,
      "grad_norm": 1.25,
      "learning_rate": 0.0004934410824269992,
      "loss": 6.3521,
      "mean_token_accuracy": 0.141788549721241,
      "num_tokens": 3310672.0,
      "step": 1820
    },
    {
      "entropy": 6.385811614990234,
      "epoch": 1.568113450795015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933611508728182,
      "loss": 6.3045,
      "mean_token_accuracy": 0.14355277717113496,
      "num_tokens": 3319666.0,
      "step": 1825
    },
    {
      "entropy": 6.430783605575561,
      "epoch": 1.5724108293940695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000493280742505162,
      "loss": 6.3168,
      "mean_token_accuracy": 0.14499156102538108,
      "num_tokens": 3329121.0,
      "step": 1830
    },
    {
      "entropy": 6.328087186813354,
      "epoch": 1.576708207993124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004931998574996102,
      "loss": 6.2982,
      "mean_token_accuracy": 0.14460441917181016,
      "num_tokens": 3337868.0,
      "step": 1835
    },
    {
      "entropy": 6.422233009338379,
      "epoch": 1.5810055865921788,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004931184960327832,
      "loss": 6.3359,
      "mean_token_accuracy": 0.14177661910653114,
      "num_tokens": 3346447.0,
      "step": 1840
    },
    {
      "entropy": 6.337698698043823,
      "epoch": 1.5853029651912334,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004930366582823421,
      "loss": 6.2971,
      "mean_token_accuracy": 0.13885951936244964,
      "num_tokens": 3355187.0,
      "step": 1845
    },
    {
      "entropy": 6.434715986251831,
      "epoch": 1.589600343790288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929543444269879,
      "loss": 6.5041,
      "mean_token_accuracy": 0.12198112457990647,
      "num_tokens": 3365978.0,
      "step": 1850
    },
    {
      "entropy": 6.421086025238037,
      "epoch": 1.5938977223893425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000492871554646461,
      "loss": 6.4553,
      "mean_token_accuracy": 0.13949850127100943,
      "num_tokens": 3374607.0,
      "step": 1855
    },
    {
      "entropy": 6.452510786056519,
      "epoch": 1.5981951009883972,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004927882891215413,
      "loss": 6.3641,
      "mean_token_accuracy": 0.14000257328152657,
      "num_tokens": 3384541.0,
      "step": 1860
    },
    {
      "entropy": 6.383066320419312,
      "epoch": 1.6024924795874518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004927045480340475,
      "loss": 6.3822,
      "mean_token_accuracy": 0.1305439554154873,
      "num_tokens": 3394009.0,
      "step": 1865
    },
    {
      "entropy": 6.446542024612427,
      "epoch": 1.6067898581865063,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004926203315668363,
      "loss": 6.3698,
      "mean_token_accuracy": 0.14797376990318298,
      "num_tokens": 3402550.0,
      "step": 1870
    },
    {
      "entropy": 6.390069961547852,
      "epoch": 1.6110872367855609,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004925356399038032,
      "loss": 6.3408,
      "mean_token_accuracy": 0.1447392597794533,
      "num_tokens": 3413230.0,
      "step": 1875
    },
    {
      "entropy": 6.27685923576355,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004924504732298808,
      "loss": 6.3117,
      "mean_token_accuracy": 0.14603087455034255,
      "num_tokens": 3422325.0,
      "step": 1880
    },
    {
      "entropy": 6.474649381637573,
      "epoch": 1.61968199398367,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004923648317310391,
      "loss": 6.392,
      "mean_token_accuracy": 0.13447986543178558,
      "num_tokens": 3431758.0,
      "step": 1885
    },
    {
      "entropy": 6.331472253799438,
      "epoch": 1.6239793725827245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922787155942849,
      "loss": 6.2769,
      "mean_token_accuracy": 0.14332059174776077,
      "num_tokens": 3441281.0,
      "step": 1890
    },
    {
      "entropy": 6.405183362960815,
      "epoch": 1.628276751181779,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004921921250076611,
      "loss": 6.32,
      "mean_token_accuracy": 0.1474598653614521,
      "num_tokens": 3449822.0,
      "step": 1895
    },
    {
      "entropy": 6.439214515686035,
      "epoch": 1.6325741297808336,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004921050601602475,
      "loss": 6.3252,
      "mean_token_accuracy": 0.14000287130475045,
      "num_tokens": 3459437.0,
      "step": 1900
    },
    {
      "entropy": 6.364074611663819,
      "epoch": 1.6368715083798882,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004920175212421587,
      "loss": 6.3453,
      "mean_token_accuracy": 0.14289369359612464,
      "num_tokens": 3469546.0,
      "step": 1905
    },
    {
      "entropy": 6.427534103393555,
      "epoch": 1.6411688869789427,
      "grad_norm": 1.125,
      "learning_rate": 0.0004919295084445445,
      "loss": 6.316,
      "mean_token_accuracy": 0.13860969692468644,
      "num_tokens": 3478521.0,
      "step": 1910
    },
    {
      "entropy": 6.320402002334594,
      "epoch": 1.6454662655779975,
      "grad_norm": 1.125,
      "learning_rate": 0.0004918410219595899,
      "loss": 6.2135,
      "mean_token_accuracy": 0.1451512724161148,
      "num_tokens": 3487958.0,
      "step": 1915
    },
    {
      "entropy": 6.288850164413452,
      "epoch": 1.649763644177052,
      "grad_norm": 1.09375,
      "learning_rate": 0.000491752061980514,
      "loss": 6.3142,
      "mean_token_accuracy": 0.14572881534695625,
      "num_tokens": 3495941.0,
      "step": 1920
    },
    {
      "entropy": 6.415158891677857,
      "epoch": 1.6540610227761066,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004916626287015697,
      "loss": 6.3295,
      "mean_token_accuracy": 0.1428305320441723,
      "num_tokens": 3505172.0,
      "step": 1925
    },
    {
      "entropy": 6.391680335998535,
      "epoch": 1.658358401375161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004915727223180436,
      "loss": 6.3138,
      "mean_token_accuracy": 0.14527018666267394,
      "num_tokens": 3514084.0,
      "step": 1930
    },
    {
      "entropy": 6.442883253097534,
      "epoch": 1.6626557799742159,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004914823430262554,
      "loss": 6.4421,
      "mean_token_accuracy": 0.13630780205130577,
      "num_tokens": 3523318.0,
      "step": 1935
    },
    {
      "entropy": 6.3719299793243405,
      "epoch": 1.6669531585732704,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913914910235573,
      "loss": 6.2712,
      "mean_token_accuracy": 0.1465964734554291,
      "num_tokens": 3531681.0,
      "step": 1940
    },
    {
      "entropy": 6.30449800491333,
      "epoch": 1.671250537172325,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004913001665083337,
      "loss": 6.3462,
      "mean_token_accuracy": 0.14266296550631524,
      "num_tokens": 3540326.0,
      "step": 1945
    },
    {
      "entropy": 6.393455266952515,
      "epoch": 1.6755479157713795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004912083696800008,
      "loss": 6.2845,
      "mean_token_accuracy": 0.13957952111959457,
      "num_tokens": 3548319.0,
      "step": 1950
    },
    {
      "entropy": 6.2641520500183105,
      "epoch": 1.679845294370434,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004911161007390063,
      "loss": 6.1394,
      "mean_token_accuracy": 0.15182972550392151,
      "num_tokens": 3557743.0,
      "step": 1955
    },
    {
      "entropy": 6.277702379226684,
      "epoch": 1.6841426729694886,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004910233598868287,
      "loss": 6.2682,
      "mean_token_accuracy": 0.14529897049069404,
      "num_tokens": 3566810.0,
      "step": 1960
    },
    {
      "entropy": 6.337201499938965,
      "epoch": 1.6884400515685432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004909301473259769,
      "loss": 6.3446,
      "mean_token_accuracy": 0.1441364496946335,
      "num_tokens": 3576273.0,
      "step": 1965
    },
    {
      "entropy": 6.315463352203369,
      "epoch": 1.6927374301675977,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004908364632599899,
      "loss": 6.3265,
      "mean_token_accuracy": 0.1417617380619049,
      "num_tokens": 3585362.0,
      "step": 1970
    },
    {
      "entropy": 6.334482049942016,
      "epoch": 1.6970348087666522,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004907423078934362,
      "loss": 6.3112,
      "mean_token_accuracy": 0.1375837117433548,
      "num_tokens": 3594445.0,
      "step": 1975
    },
    {
      "entropy": 6.350462293624878,
      "epoch": 1.7013321873657068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004906476814319134,
      "loss": 6.3339,
      "mean_token_accuracy": 0.14064698591828345,
      "num_tokens": 3604046.0,
      "step": 1980
    },
    {
      "entropy": 6.426722812652588,
      "epoch": 1.7056295659647613,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905525840820481,
      "loss": 6.3385,
      "mean_token_accuracy": 0.14573438987135887,
      "num_tokens": 3613264.0,
      "step": 1985
    },
    {
      "entropy": 6.352614307403565,
      "epoch": 1.709926944563816,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004904570160514948,
      "loss": 6.2274,
      "mean_token_accuracy": 0.1472587391734123,
      "num_tokens": 3620938.0,
      "step": 1990
    },
    {
      "entropy": 6.332766199111939,
      "epoch": 1.7142243231628707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004903609775489358,
      "loss": 6.3316,
      "mean_token_accuracy": 0.141658017039299,
      "num_tokens": 3629782.0,
      "step": 1995
    },
    {
      "entropy": 6.470893049240113,
      "epoch": 1.7185217017619252,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004902644687840809,
      "loss": 6.3851,
      "mean_token_accuracy": 0.13716112747788428,
      "num_tokens": 3638389.0,
      "step": 2000
    },
    {
      "epoch": 1.7185217017619252,
      "eval_entropy": 6.099528554323557,
      "eval_loss": 6.394359588623047,
      "eval_mean_token_accuracy": 0.1416532844808456,
      "eval_num_tokens": 3638389.0,
      "eval_runtime": 2.6474,
      "eval_samples_per_second": 1340.554,
      "eval_steps_per_second": 167.711,
      "step": 2000
    },
    {
      "entropy": 6.261544561386108,
      "epoch": 1.7228190803609797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004901674899676667,
      "loss": 6.2108,
      "mean_token_accuracy": 0.15050435364246367,
      "num_tokens": 3647196.0,
      "step": 2005
    },
    {
      "entropy": 6.274995756149292,
      "epoch": 1.7271164589600345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004900700413114561,
      "loss": 6.1529,
      "mean_token_accuracy": 0.14831772446632385,
      "num_tokens": 3655931.0,
      "step": 2010
    },
    {
      "entropy": 6.3229820728302,
      "epoch": 1.731413837559089,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489972123028238,
      "loss": 6.321,
      "mean_token_accuracy": 0.14567042738199235,
      "num_tokens": 3665340.0,
      "step": 2015
    },
    {
      "entropy": 6.3659852027893065,
      "epoch": 1.7357112161581436,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004898737353318268,
      "loss": 6.2768,
      "mean_token_accuracy": 0.1477080449461937,
      "num_tokens": 3674674.0,
      "step": 2020
    },
    {
      "entropy": 6.348592519760132,
      "epoch": 1.7400085947571982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489774878437062,
      "loss": 6.3145,
      "mean_token_accuracy": 0.14947012513875962,
      "num_tokens": 3683910.0,
      "step": 2025
    },
    {
      "entropy": 6.259040975570679,
      "epoch": 1.7443059733562527,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004896755525598074,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14689078107476233,
      "num_tokens": 3692772.0,
      "step": 2030
    },
    {
      "entropy": 6.342519044876099,
      "epoch": 1.7486033519553073,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895757579169511,
      "loss": 6.2427,
      "mean_token_accuracy": 0.15017979145050048,
      "num_tokens": 3701875.0,
      "step": 2035
    },
    {
      "entropy": 6.292976188659668,
      "epoch": 1.7529007305543618,
      "grad_norm": 1.125,
      "learning_rate": 0.0004894754947264047,
      "loss": 6.2553,
      "mean_token_accuracy": 0.13835104927420616,
      "num_tokens": 3711059.0,
      "step": 2040
    },
    {
      "entropy": 6.299735260009766,
      "epoch": 1.7571981091534163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489374763207103,
      "loss": 6.2281,
      "mean_token_accuracy": 0.15146950185298919,
      "num_tokens": 3719667.0,
      "step": 2045
    },
    {
      "entropy": 6.301810169219971,
      "epoch": 1.761495487752471,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004892735635790033,
      "loss": 6.2123,
      "mean_token_accuracy": 0.1495981313288212,
      "num_tokens": 3728905.0,
      "step": 2050
    },
    {
      "entropy": 6.30145845413208,
      "epoch": 1.7657928663515254,
      "grad_norm": 1.078125,
      "learning_rate": 0.000489171896063085,
      "loss": 6.2537,
      "mean_token_accuracy": 0.1459835670888424,
      "num_tokens": 3737973.0,
      "step": 2055
    },
    {
      "entropy": 6.2883388042449955,
      "epoch": 1.77009024495058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004890697608813495,
      "loss": 6.2163,
      "mean_token_accuracy": 0.14961224421858788,
      "num_tokens": 3747160.0,
      "step": 2060
    },
    {
      "entropy": 6.263640260696411,
      "epoch": 1.7743876235496348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004889671582568193,
      "loss": 6.1784,
      "mean_token_accuracy": 0.15031145811080932,
      "num_tokens": 3754908.0,
      "step": 2065
    },
    {
      "entropy": 6.346582460403442,
      "epoch": 1.7786850021486893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004888640884135374,
      "loss": 6.2829,
      "mean_token_accuracy": 0.13900179788470268,
      "num_tokens": 3763588.0,
      "step": 2070
    },
    {
      "entropy": 6.257235670089722,
      "epoch": 1.7829823807477438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004887605515765671,
      "loss": 6.1709,
      "mean_token_accuracy": 0.14929060488939286,
      "num_tokens": 3771932.0,
      "step": 2075
    },
    {
      "entropy": 6.365653419494629,
      "epoch": 1.7872797593467986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004886565479719914,
      "loss": 6.3345,
      "mean_token_accuracy": 0.14113251492381096,
      "num_tokens": 3780761.0,
      "step": 2080
    },
    {
      "entropy": 6.262348651885986,
      "epoch": 1.7915771379458532,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004885520778269128,
      "loss": 6.1321,
      "mean_token_accuracy": 0.1545582115650177,
      "num_tokens": 3789511.0,
      "step": 2085
    },
    {
      "entropy": 6.2817658424377445,
      "epoch": 1.7958745165449077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004884471413694523,
      "loss": 6.314,
      "mean_token_accuracy": 0.13621936738491058,
      "num_tokens": 3799068.0,
      "step": 2090
    },
    {
      "entropy": 6.329490661621094,
      "epoch": 1.8001718951439623,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004883417388287491,
      "loss": 6.203,
      "mean_token_accuracy": 0.14342657700181008,
      "num_tokens": 3807850.0,
      "step": 2095
    },
    {
      "entropy": 6.38734540939331,
      "epoch": 1.8044692737430168,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004882358704349603,
      "loss": 6.2334,
      "mean_token_accuracy": 0.1539076805114746,
      "num_tokens": 3816232.0,
      "step": 2100
    },
    {
      "entropy": 6.277137184143067,
      "epoch": 1.8087666523420713,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004881295364192601,
      "loss": 6.2852,
      "mean_token_accuracy": 0.14138692021369934,
      "num_tokens": 3825334.0,
      "step": 2105
    },
    {
      "entropy": 6.2631055355072025,
      "epoch": 1.813064030941126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004880227370138394,
      "loss": 6.249,
      "mean_token_accuracy": 0.1496968075633049,
      "num_tokens": 3834756.0,
      "step": 2110
    },
    {
      "entropy": 6.390051364898682,
      "epoch": 1.8173614095401804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004879154724519057,
      "loss": 6.3121,
      "mean_token_accuracy": 0.14131874218583107,
      "num_tokens": 3843852.0,
      "step": 2115
    },
    {
      "entropy": 6.326163721084595,
      "epoch": 1.821658788139235,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004878077429676816,
      "loss": 6.1782,
      "mean_token_accuracy": 0.15569255948066713,
      "num_tokens": 3852213.0,
      "step": 2120
    },
    {
      "entropy": 6.302305793762207,
      "epoch": 1.8259561667382895,
      "grad_norm": 1.125,
      "learning_rate": 0.0004876995487964054,
      "loss": 6.3249,
      "mean_token_accuracy": 0.13407231569290162,
      "num_tokens": 3861522.0,
      "step": 2125
    },
    {
      "entropy": 6.268641996383667,
      "epoch": 1.830253545337344,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048759089017432996,
      "loss": 6.2752,
      "mean_token_accuracy": 0.1495513841509819,
      "num_tokens": 3870328.0,
      "step": 2130
    },
    {
      "entropy": 6.366842555999756,
      "epoch": 1.8345509239363988,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004874817673387222,
      "loss": 6.253,
      "mean_token_accuracy": 0.14391155242919923,
      "num_tokens": 3880399.0,
      "step": 2135
    },
    {
      "entropy": 6.331115055084228,
      "epoch": 1.8388483025354534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048737218052786275,
      "loss": 6.202,
      "mean_token_accuracy": 0.146192529797554,
      "num_tokens": 3889859.0,
      "step": 2140
    },
    {
      "entropy": 6.224690389633179,
      "epoch": 1.843145681134508,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048726212998104554,
      "loss": 6.1712,
      "mean_token_accuracy": 0.1507162019610405,
      "num_tokens": 3899609.0,
      "step": 2145
    },
    {
      "entropy": 6.3443187236785885,
      "epoch": 1.8474430597335625,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004871516159385768,
      "loss": 6.2796,
      "mean_token_accuracy": 0.13646257296204567,
      "num_tokens": 3909452.0,
      "step": 2150
    },
    {
      "entropy": 6.383110523223877,
      "epoch": 1.8517404383326173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004870406386417752,
      "loss": 6.3188,
      "mean_token_accuracy": 0.14102417081594468,
      "num_tokens": 3918579.0,
      "step": 2155
    },
    {
      "entropy": 6.2510617733001705,
      "epoch": 1.8560378169316718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869291983329707,
      "loss": 6.2026,
      "mean_token_accuracy": 0.15068159848451615,
      "num_tokens": 3926951.0,
      "step": 2160
    },
    {
      "entropy": 6.326785755157471,
      "epoch": 1.8603351955307263,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004868172952555044,
      "loss": 6.2538,
      "mean_token_accuracy": 0.14530059248208999,
      "num_tokens": 3935771.0,
      "step": 2165
    },
    {
      "entropy": 6.247272396087647,
      "epoch": 1.864632574129781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004867049296537278,
      "loss": 6.1507,
      "mean_token_accuracy": 0.15399375408887864,
      "num_tokens": 3944577.0,
      "step": 2170
    },
    {
      "entropy": 6.3375420570373535,
      "epoch": 1.8689299527288354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004865921017730027,
      "loss": 6.3118,
      "mean_token_accuracy": 0.14250392094254494,
      "num_tokens": 3954499.0,
      "step": 2175
    },
    {
      "entropy": 6.286665630340576,
      "epoch": 1.87322733132789,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048647881185969995,
      "loss": 6.2527,
      "mean_token_accuracy": 0.14288917630910875,
      "num_tokens": 3963953.0,
      "step": 2180
    },
    {
      "entropy": 6.279871034622192,
      "epoch": 1.8775247099269445,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004863650601611994,
      "loss": 6.2466,
      "mean_token_accuracy": 0.14417033568024634,
      "num_tokens": 3973925.0,
      "step": 2185
    },
    {
      "entropy": 6.310920095443725,
      "epoch": 1.881822088525999,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048625084692588937,
      "loss": 6.2296,
      "mean_token_accuracy": 0.1480042040348053,
      "num_tokens": 3983280.0,
      "step": 2190
    },
    {
      "entropy": 6.281775856018067,
      "epoch": 1.8861194671250536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048613617240316593,
      "loss": 6.2275,
      "mean_token_accuracy": 0.14622583016753196,
      "num_tokens": 3992767.0,
      "step": 2195
    },
    {
      "entropy": 6.342629909515381,
      "epoch": 1.8904168457241082,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004860210368434323,
      "loss": 6.3037,
      "mean_token_accuracy": 0.1325987122952938,
      "num_tokens": 4001600.0,
      "step": 2200
    },
    {
      "entropy": 6.265765047073364,
      "epoch": 1.8947142243231627,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048590544049809857,
      "loss": 6.2767,
      "mean_token_accuracy": 0.1438748598098755,
      "num_tokens": 4010852.0,
      "step": 2205
    },
    {
      "entropy": 6.340087509155273,
      "epoch": 1.8990116029222175,
      "grad_norm": 1.015625,
      "learning_rate": 0.000485789383619581,
      "loss": 6.1808,
      "mean_token_accuracy": 0.1515198156237602,
      "num_tokens": 4020280.0,
      "step": 2210
    },
    {
      "entropy": 6.251649570465088,
      "epoch": 1.903308981521272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004856728664613015,
      "loss": 6.1924,
      "mean_token_accuracy": 0.14853738620877266,
      "num_tokens": 4029283.0,
      "step": 2215
    },
    {
      "entropy": 6.3928605079650875,
      "epoch": 1.9076063601203266,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048555588927768674,
      "loss": 6.2813,
      "mean_token_accuracy": 0.15265752896666526,
      "num_tokens": 4038704.0,
      "step": 2220
    },
    {
      "entropy": 6.133363246917725,
      "epoch": 1.9119037387193811,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004854384523241683,
      "loss": 6.149,
      "mean_token_accuracy": 0.14911998957395553,
      "num_tokens": 4047492.0,
      "step": 2225
    },
    {
      "entropy": 6.350856971740723,
      "epoch": 1.916201117318436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048532055585718143,
      "loss": 6.3258,
      "mean_token_accuracy": 0.14355307817459106,
      "num_tokens": 4056792.0,
      "step": 2230
    },
    {
      "entropy": 6.365505075454712,
      "epoch": 1.9204984959174904,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048520220013416505,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1529555104672909,
      "num_tokens": 4065284.0,
      "step": 2235
    },
    {
      "entropy": 6.168982744216919,
      "epoch": 1.924795874516545,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004850833854135607,
      "loss": 6.2088,
      "mean_token_accuracy": 0.1461514100432396,
      "num_tokens": 4074727.0,
      "step": 2240
    },
    {
      "entropy": 6.41944990158081,
      "epoch": 1.9290932531155995,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004849641119548122,
      "loss": 6.2592,
      "mean_token_accuracy": 0.15032583549618722,
      "num_tokens": 4083360.0,
      "step": 2245
    },
    {
      "entropy": 6.2758283615112305,
      "epoch": 1.933390631714654,
      "grad_norm": 1.203125,
      "learning_rate": 0.000484844380018365,
      "loss": 6.3513,
      "mean_token_accuracy": 0.13633401170372964,
      "num_tokens": 4092431.0,
      "step": 2250
    },
    {
      "entropy": 6.3127405643463135,
      "epoch": 1.9376880103137086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000484724189865666,
      "loss": 6.2257,
      "mean_token_accuracy": 0.1493070751428604,
      "num_tokens": 4102198.0,
      "step": 2255
    },
    {
      "entropy": 6.309014892578125,
      "epoch": 1.9419853889127632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004846035417591624,
      "loss": 6.2159,
      "mean_token_accuracy": 0.14939168095588684,
      "num_tokens": 4111568.0,
      "step": 2260
    },
    {
      "entropy": 6.260059499740601,
      "epoch": 1.9462827675118177,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004844824359623014,
      "loss": 6.2428,
      "mean_token_accuracy": 0.1440415069460869,
      "num_tokens": 4121151.0,
      "step": 2265
    },
    {
      "entropy": 6.318060827255249,
      "epoch": 1.9505801461108723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048436087273952966,
      "loss": 6.1757,
      "mean_token_accuracy": 0.14666527807712554,
      "num_tokens": 4130197.0,
      "step": 2270
    },
    {
      "entropy": 6.161466360092163,
      "epoch": 1.9548775247099268,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048423885235629265,
      "loss": 6.1784,
      "mean_token_accuracy": 0.1395201563835144,
      "num_tokens": 4138310.0,
      "step": 2275
    },
    {
      "entropy": 6.2752457618713375,
      "epoch": 1.9591749033089814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004841163750790342,
      "loss": 6.1845,
      "mean_token_accuracy": 0.15314925760030745,
      "num_tokens": 4148270.0,
      "step": 2280
    },
    {
      "entropy": 6.315674591064453,
      "epoch": 1.9634722819080361,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048399344117519555,
      "loss": 6.2468,
      "mean_token_accuracy": 0.14707700312137603,
      "num_tokens": 4157276.0,
      "step": 2285
    },
    {
      "entropy": 6.204446601867676,
      "epoch": 1.9677696605070907,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048387005091321544,
      "loss": 6.2742,
      "mean_token_accuracy": 0.15154237672686577,
      "num_tokens": 4166144.0,
      "step": 2290
    },
    {
      "entropy": 6.408171319961548,
      "epoch": 1.9720670391061452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048374620456252877,
      "loss": 6.2501,
      "mean_token_accuracy": 0.14817840680480004,
      "num_tokens": 4175146.0,
      "step": 2295
    },
    {
      "entropy": 6.269484567642212,
      "epoch": 1.9763644177052,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048362190239356644,
      "loss": 6.2973,
      "mean_token_accuracy": 0.14276304692029954,
      "num_tokens": 4184089.0,
      "step": 2300
    },
    {
      "entropy": 6.2955399513244625,
      "epoch": 1.9806617963042545,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048349714467775474,
      "loss": 6.2199,
      "mean_token_accuracy": 0.14973291605710984,
      "num_tokens": 4192453.0,
      "step": 2305
    },
    {
      "entropy": 6.279147005081176,
      "epoch": 1.984959174903309,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048337193168751464,
      "loss": 6.1458,
      "mean_token_accuracy": 0.15528891682624818,
      "num_tokens": 4201323.0,
      "step": 2310
    },
    {
      "entropy": 6.252904844284058,
      "epoch": 1.9892565535023636,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004832462636962613,
      "loss": 6.1965,
      "mean_token_accuracy": 0.14820378869771958,
      "num_tokens": 4209805.0,
      "step": 2315
    },
    {
      "entropy": 6.274013090133667,
      "epoch": 1.9935539321014182,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004831201409784034,
      "loss": 6.1716,
      "mean_token_accuracy": 0.1535733938217163,
      "num_tokens": 4218562.0,
      "step": 2320
    },
    {
      "entropy": 6.276880836486816,
      "epoch": 1.9978513107004727,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004829935638093424,
      "loss": 6.2489,
      "mean_token_accuracy": 0.14822545498609543,
      "num_tokens": 4227605.0,
      "step": 2325
    },
    {
      "entropy": 6.305307547251384,
      "epoch": 2.0017189514396216,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004828665324654724,
      "loss": 6.0808,
      "mean_token_accuracy": 0.15109746985965306,
      "num_tokens": 4235441.0,
      "step": 2330
    },
    {
      "entropy": 6.1959953784942625,
      "epoch": 2.006016330038676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004827390472241791,
      "loss": 5.8773,
      "mean_token_accuracy": 0.15755202025175094,
      "num_tokens": 4243982.0,
      "step": 2335
    },
    {
      "entropy": 6.151334667205811,
      "epoch": 2.010313708637731,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004826111083638392,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15486665964126586,
      "num_tokens": 4253032.0,
      "step": 2340
    },
    {
      "entropy": 6.237190914154053,
      "epoch": 2.0146110872367857,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048248271616382,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1612431138753891,
      "num_tokens": 4262000.0,
      "step": 2345
    },
    {
      "entropy": 6.14679217338562,
      "epoch": 2.0189084658358403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048235387090447894,
      "loss": 5.8772,
      "mean_token_accuracy": 0.16036980897188186,
      "num_tokens": 4271089.0,
      "step": 2350
    },
    {
      "entropy": 6.216695976257324,
      "epoch": 2.023205844434895,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048222457286716235,
      "loss": 5.8748,
      "mean_token_accuracy": 0.1604112282395363,
      "num_tokens": 4280064.0,
      "step": 2355
    },
    {
      "entropy": 6.1386841297149655,
      "epoch": 2.0275032230339494,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048209482233420564,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16113310903310776,
      "num_tokens": 4289096.0,
      "step": 2360
    },
    {
      "entropy": 6.134747457504273,
      "epoch": 2.031800601633004,
      "grad_norm": 1.140625,
      "learning_rate": 0.000481964619588932,
      "loss": 5.878,
      "mean_token_accuracy": 0.1504762887954712,
      "num_tokens": 4298341.0,
      "step": 2365
    },
    {
      "entropy": 6.1863165378570555,
      "epoch": 2.0360979802320585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004818339649156523,
      "loss": 5.927,
      "mean_token_accuracy": 0.1504568099975586,
      "num_tokens": 4307466.0,
      "step": 2370
    },
    {
      "entropy": 6.122588872909546,
      "epoch": 2.040395358831113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048170285859966395,
      "loss": 5.8523,
      "mean_token_accuracy": 0.15609132051467894,
      "num_tokens": 4316559.0,
      "step": 2375
    },
    {
      "entropy": 6.180559778213501,
      "epoch": 2.0446927374301676,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048157130092725087,
      "loss": 5.8607,
      "mean_token_accuracy": 0.1605754405260086,
      "num_tokens": 4326002.0,
      "step": 2380
    },
    {
      "entropy": 6.1395592212677,
      "epoch": 2.048990116029222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004814392921856824,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16454231590032578,
      "num_tokens": 4333767.0,
      "step": 2385
    },
    {
      "entropy": 6.13576455116272,
      "epoch": 2.0532874946282766,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004813068326632128,
      "loss": 5.8941,
      "mean_token_accuracy": 0.16491636782884597,
      "num_tokens": 4342726.0,
      "step": 2390
    },
    {
      "entropy": 6.0936521053314205,
      "epoch": 2.057584873227331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004811739226490809,
      "loss": 5.7842,
      "mean_token_accuracy": 0.16501135975122452,
      "num_tokens": 4351278.0,
      "step": 2395
    },
    {
      "entropy": 6.10044412612915,
      "epoch": 2.0618822518263857,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048104056243350896,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16724471151828765,
      "num_tokens": 4360146.0,
      "step": 2400
    },
    {
      "entropy": 6.106970643997192,
      "epoch": 2.0661796304254403,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004809067523077023,
      "loss": 5.8729,
      "mean_token_accuracy": 0.157526695728302,
      "num_tokens": 4368474.0,
      "step": 2405
    },
    {
      "entropy": 6.171187162399292,
      "epoch": 2.0704770090244953,
      "grad_norm": 1.375,
      "learning_rate": 0.00048077249256384884,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1573039174079895,
      "num_tokens": 4378430.0,
      "step": 2410
    },
    {
      "entropy": 6.091280841827393,
      "epoch": 2.07477438762355,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004806377834951182,
      "loss": 5.935,
      "mean_token_accuracy": 0.16114620715379716,
      "num_tokens": 4387661.0,
      "step": 2415
    },
    {
      "entropy": 6.2243939399719235,
      "epoch": 2.0790717662226044,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048050262539566104,
      "loss": 5.878,
      "mean_token_accuracy": 0.16368078142404557,
      "num_tokens": 4396568.0,
      "step": 2420
    },
    {
      "entropy": 6.03709397315979,
      "epoch": 2.083369144821659,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004803670185606087,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16086018234491348,
      "num_tokens": 4405113.0,
      "step": 2425
    },
    {
      "entropy": 6.130343770980835,
      "epoch": 2.0876665234207135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004802309632860724,
      "loss": 5.9079,
      "mean_token_accuracy": 0.15909243375062943,
      "num_tokens": 4414930.0,
      "step": 2430
    },
    {
      "entropy": 6.135715103149414,
      "epoch": 2.091963902019768,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048009445986914236,
      "loss": 5.9472,
      "mean_token_accuracy": 0.15476218312978746,
      "num_tokens": 4424200.0,
      "step": 2435
    },
    {
      "entropy": 6.087995195388794,
      "epoch": 2.0962612806188226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047995750860788756,
      "loss": 5.8909,
      "mean_token_accuracy": 0.16245823204517365,
      "num_tokens": 4433533.0,
      "step": 2440
    },
    {
      "entropy": 6.122352218627929,
      "epoch": 2.100558659217877,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004798201098013547,
      "loss": 5.9404,
      "mean_token_accuracy": 0.1550127923488617,
      "num_tokens": 4442752.0,
      "step": 2445
    },
    {
      "entropy": 6.078053426742554,
      "epoch": 2.1048560378169316,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047968226374956797,
      "loss": 5.971,
      "mean_token_accuracy": 0.15189971923828124,
      "num_tokens": 4453624.0,
      "step": 2450
    },
    {
      "entropy": 6.135152006149292,
      "epoch": 2.109153416415986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047954397075352794,
      "loss": 5.9528,
      "mean_token_accuracy": 0.15271927714347838,
      "num_tokens": 4462101.0,
      "step": 2455
    },
    {
      "entropy": 6.186189222335815,
      "epoch": 2.1134507950150407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047940523111521136,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15061386451125144,
      "num_tokens": 4471274.0,
      "step": 2460
    },
    {
      "entropy": 6.176623678207397,
      "epoch": 2.1177481736140953,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004792660451375701,
      "loss": 5.9378,
      "mean_token_accuracy": 0.15836917012929916,
      "num_tokens": 4480281.0,
      "step": 2465
    },
    {
      "entropy": 6.131458759307861,
      "epoch": 2.12204555221315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047912641312453064,
      "loss": 5.8644,
      "mean_token_accuracy": 0.15682227164506912,
      "num_tokens": 4489325.0,
      "step": 2470
    },
    {
      "entropy": 6.136132478713989,
      "epoch": 2.1263429308122044,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047898633538099363,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16200080662965774,
      "num_tokens": 4497977.0,
      "step": 2475
    },
    {
      "entropy": 6.082278108596801,
      "epoch": 2.130640309411259,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004788458122128327,
      "loss": 5.9842,
      "mean_token_accuracy": 0.15614993274211883,
      "num_tokens": 4509133.0,
      "step": 2480
    },
    {
      "entropy": 6.15927677154541,
      "epoch": 2.134937688010314,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047870484392689434,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16319320052862168,
      "num_tokens": 4518431.0,
      "step": 2485
    },
    {
      "entropy": 6.090216445922851,
      "epoch": 2.1392350666093685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000478563430830997,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15791197270154952,
      "num_tokens": 4528489.0,
      "step": 2490
    },
    {
      "entropy": 6.161773586273194,
      "epoch": 2.143532445208423,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047842157323393035,
      "loss": 5.821,
      "mean_token_accuracy": 0.1565369412302971,
      "num_tokens": 4537068.0,
      "step": 2495
    },
    {
      "entropy": 6.096503257751465,
      "epoch": 2.1478298238074776,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004782792714454547,
      "loss": 5.9322,
      "mean_token_accuracy": 0.16345749497413636,
      "num_tokens": 4545773.0,
      "step": 2500
    },
    {
      "epoch": 2.1478298238074776,
      "eval_entropy": 5.961080405089232,
      "eval_loss": 6.284891128540039,
      "eval_mean_token_accuracy": 0.152339708798372,
      "eval_num_tokens": 4545773.0,
      "eval_runtime": 2.6434,
      "eval_samples_per_second": 1342.578,
      "eval_steps_per_second": 167.964,
      "step": 2500
    }
  ],
  "logging_steps": 5,
  "max_steps": 11630,
  "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": 6872272353792000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}