{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0242057488653553,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.910352373123169,
      "epoch": 0.005042864346949067,
      "grad_norm": 21.125,
      "learning_rate": 2e-06,
      "loss": 14.2192,
      "mean_token_accuracy": 0.0,
      "num_tokens": 11445.0,
      "step": 5
    },
    {
      "entropy": 4.904936599731445,
      "epoch": 0.010085728693898134,
      "grad_norm": 21.875,
      "learning_rate": 4.5e-06,
      "loss": 14.1143,
      "mean_token_accuracy": 6.73400645609945e-05,
      "num_tokens": 22095.0,
      "step": 10
    },
    {
      "entropy": 4.916885423660278,
      "epoch": 0.015128593040847202,
      "grad_norm": 22.875,
      "learning_rate": 7e-06,
      "loss": 14.1034,
      "mean_token_accuracy": 0.00013709709164686502,
      "num_tokens": 33847.0,
      "step": 15
    },
    {
      "entropy": 5.0439094543457035,
      "epoch": 0.020171457387796268,
      "grad_norm": 27.75,
      "learning_rate": 9.5e-06,
      "loss": 13.7381,
      "mean_token_accuracy": 0.0,
      "num_tokens": 44495.0,
      "step": 20
    },
    {
      "entropy": 5.412609577178955,
      "epoch": 0.025214321734745335,
      "grad_norm": 37.25,
      "learning_rate": 1.2e-05,
      "loss": 13.0504,
      "mean_token_accuracy": 0.0,
      "num_tokens": 55298.0,
      "step": 25
    },
    {
      "entropy": 6.753208303451538,
      "epoch": 0.030257186081694403,
      "grad_norm": 31.25,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 12.0538,
      "mean_token_accuracy": 0.0,
      "num_tokens": 66391.0,
      "step": 30
    },
    {
      "entropy": 9.891419410705566,
      "epoch": 0.03530005042864347,
      "grad_norm": 3.484375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.9607,
      "mean_token_accuracy": 0.00013166556600481273,
      "num_tokens": 77624.0,
      "step": 35
    },
    {
      "entropy": 10.698719787597657,
      "epoch": 0.040342914775592535,
      "grad_norm": 3.5625,
      "learning_rate": 1.95e-05,
      "loss": 10.7344,
      "mean_token_accuracy": 0.00032866199035197496,
      "num_tokens": 89426.0,
      "step": 40
    },
    {
      "entropy": 10.726075839996337,
      "epoch": 0.0453857791225416,
      "grad_norm": 2.8125,
      "learning_rate": 2.2e-05,
      "loss": 10.5926,
      "mean_token_accuracy": 0.00417199133662507,
      "num_tokens": 100252.0,
      "step": 45
    },
    {
      "entropy": 10.712922477722168,
      "epoch": 0.05042864346949067,
      "grad_norm": 2.328125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.3898,
      "mean_token_accuracy": 0.014383823238313198,
      "num_tokens": 110578.0,
      "step": 50
    },
    {
      "entropy": 10.644971942901611,
      "epoch": 0.05547150781643974,
      "grad_norm": 2.265625,
      "learning_rate": 2.7e-05,
      "loss": 10.2471,
      "mean_token_accuracy": 0.015645577292889356,
      "num_tokens": 121631.0,
      "step": 55
    },
    {
      "entropy": 10.69424057006836,
      "epoch": 0.060514372163388806,
      "grad_norm": 2.078125,
      "learning_rate": 2.95e-05,
      "loss": 10.1162,
      "mean_token_accuracy": 0.01469698864966631,
      "num_tokens": 134254.0,
      "step": 60
    },
    {
      "entropy": 10.732514667510987,
      "epoch": 0.06555723651033787,
      "grad_norm": 1.6796875,
      "learning_rate": 3.2e-05,
      "loss": 10.0187,
      "mean_token_accuracy": 0.021757309418171646,
      "num_tokens": 144928.0,
      "step": 65
    },
    {
      "entropy": 10.716106986999511,
      "epoch": 0.07060010085728693,
      "grad_norm": 1.8046875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8923,
      "mean_token_accuracy": 0.03248250037431717,
      "num_tokens": 155518.0,
      "step": 70
    },
    {
      "entropy": 10.709668159484863,
      "epoch": 0.07564296520423601,
      "grad_norm": 1.8984375,
      "learning_rate": 3.7e-05,
      "loss": 9.8081,
      "mean_token_accuracy": 0.032918563671410085,
      "num_tokens": 165850.0,
      "step": 75
    },
    {
      "entropy": 10.714706993103027,
      "epoch": 0.08068582955118507,
      "grad_norm": 1.8359375,
      "learning_rate": 3.95e-05,
      "loss": 9.7875,
      "mean_token_accuracy": 0.03219325188547373,
      "num_tokens": 176787.0,
      "step": 80
    },
    {
      "entropy": 10.714124965667725,
      "epoch": 0.08572869389813415,
      "grad_norm": 1.7265625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.7108,
      "mean_token_accuracy": 0.035928618907928464,
      "num_tokens": 188345.0,
      "step": 85
    },
    {
      "entropy": 10.705197811126709,
      "epoch": 0.0907715582450832,
      "grad_norm": 1.7421875,
      "learning_rate": 4.45e-05,
      "loss": 9.6412,
      "mean_token_accuracy": 0.0326865192502737,
      "num_tokens": 199315.0,
      "step": 90
    },
    {
      "entropy": 10.699298572540282,
      "epoch": 0.09581442259203228,
      "grad_norm": 1.6875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4768,
      "mean_token_accuracy": 0.03807435855269432,
      "num_tokens": 209627.0,
      "step": 95
    },
    {
      "entropy": 10.682019710540771,
      "epoch": 0.10085728693898134,
      "grad_norm": 1.8046875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3915,
      "mean_token_accuracy": 0.036197875253856183,
      "num_tokens": 220872.0,
      "step": 100
    },
    {
      "entropy": 10.66194772720337,
      "epoch": 0.1059001512859304,
      "grad_norm": 1.7890625,
      "learning_rate": 5.2e-05,
      "loss": 9.2819,
      "mean_token_accuracy": 0.04516553804278374,
      "num_tokens": 233272.0,
      "step": 105
    },
    {
      "entropy": 10.633659744262696,
      "epoch": 0.11094301563287948,
      "grad_norm": 1.765625,
      "learning_rate": 5.45e-05,
      "loss": 9.1755,
      "mean_token_accuracy": 0.05018138997256756,
      "num_tokens": 243529.0,
      "step": 110
    },
    {
      "entropy": 10.592283630371094,
      "epoch": 0.11598587997982854,
      "grad_norm": 1.609375,
      "learning_rate": 5.7e-05,
      "loss": 9.0954,
      "mean_token_accuracy": 0.04114586636424065,
      "num_tokens": 254942.0,
      "step": 115
    },
    {
      "entropy": 10.565635299682617,
      "epoch": 0.12102874432677761,
      "grad_norm": 1.6796875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9839,
      "mean_token_accuracy": 0.048531080782413485,
      "num_tokens": 266552.0,
      "step": 120
    },
    {
      "entropy": 10.520887851715088,
      "epoch": 0.1260716086737267,
      "grad_norm": 1.546875,
      "learning_rate": 6.2e-05,
      "loss": 8.8489,
      "mean_token_accuracy": 0.048508816957473756,
      "num_tokens": 277906.0,
      "step": 125
    },
    {
      "entropy": 10.44383192062378,
      "epoch": 0.13111447302067575,
      "grad_norm": 1.59375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7307,
      "mean_token_accuracy": 0.05025425441563129,
      "num_tokens": 289079.0,
      "step": 130
    },
    {
      "entropy": 10.368312740325928,
      "epoch": 0.1361573373676248,
      "grad_norm": 1.4375,
      "learning_rate": 6.7e-05,
      "loss": 8.6518,
      "mean_token_accuracy": 0.04377192948013544,
      "num_tokens": 300107.0,
      "step": 135
    },
    {
      "entropy": 10.284347629547119,
      "epoch": 0.14120020171457387,
      "grad_norm": 1.46875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4993,
      "mean_token_accuracy": 0.050150758028030394,
      "num_tokens": 310779.0,
      "step": 140
    },
    {
      "entropy": 10.191308975219727,
      "epoch": 0.14624306606152296,
      "grad_norm": 1.453125,
      "learning_rate": 7.2e-05,
      "loss": 8.453,
      "mean_token_accuracy": 0.04963008016347885,
      "num_tokens": 323478.0,
      "step": 145
    },
    {
      "entropy": 10.073376941680909,
      "epoch": 0.15128593040847202,
      "grad_norm": 1.296875,
      "learning_rate": 7.45e-05,
      "loss": 8.2947,
      "mean_token_accuracy": 0.052390134707093236,
      "num_tokens": 335255.0,
      "step": 150
    },
    {
      "entropy": 9.928036785125732,
      "epoch": 0.15632879475542108,
      "grad_norm": 1.25,
      "learning_rate": 7.7e-05,
      "loss": 8.1781,
      "mean_token_accuracy": 0.050037645548582074,
      "num_tokens": 344824.0,
      "step": 155
    },
    {
      "entropy": 9.778972053527832,
      "epoch": 0.16137165910237014,
      "grad_norm": 1.140625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.1288,
      "mean_token_accuracy": 0.04721053130924702,
      "num_tokens": 356283.0,
      "step": 160
    },
    {
      "entropy": 9.597450828552246,
      "epoch": 0.1664145234493192,
      "grad_norm": 0.9140625,
      "learning_rate": 8.2e-05,
      "loss": 8.0566,
      "mean_token_accuracy": 0.04783501923084259,
      "num_tokens": 368098.0,
      "step": 165
    },
    {
      "entropy": 9.383034706115723,
      "epoch": 0.1714573877962683,
      "grad_norm": 0.94140625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.9436,
      "mean_token_accuracy": 0.048526646196842195,
      "num_tokens": 380948.0,
      "step": 170
    },
    {
      "entropy": 9.10937566757202,
      "epoch": 0.17650025214321735,
      "grad_norm": 0.87890625,
      "learning_rate": 8.7e-05,
      "loss": 7.8313,
      "mean_token_accuracy": 0.050237832218408586,
      "num_tokens": 390392.0,
      "step": 175
    },
    {
      "entropy": 8.893278980255127,
      "epoch": 0.1815431164901664,
      "grad_norm": 0.953125,
      "learning_rate": 8.95e-05,
      "loss": 7.8307,
      "mean_token_accuracy": 0.050915177911520004,
      "num_tokens": 400328.0,
      "step": 180
    },
    {
      "entropy": 8.709152698516846,
      "epoch": 0.18658598083711547,
      "grad_norm": 0.71875,
      "learning_rate": 9.2e-05,
      "loss": 7.8375,
      "mean_token_accuracy": 0.04621901027858257,
      "num_tokens": 410909.0,
      "step": 185
    },
    {
      "entropy": 8.523740673065186,
      "epoch": 0.19162884518406456,
      "grad_norm": 0.58984375,
      "learning_rate": 9.45e-05,
      "loss": 7.7759,
      "mean_token_accuracy": 0.047038092836737634,
      "num_tokens": 422726.0,
      "step": 190
    },
    {
      "entropy": 8.369872570037842,
      "epoch": 0.19667170953101362,
      "grad_norm": 0.69921875,
      "learning_rate": 9.7e-05,
      "loss": 7.6757,
      "mean_token_accuracy": 0.05148980468511581,
      "num_tokens": 433376.0,
      "step": 195
    },
    {
      "entropy": 8.261030673980713,
      "epoch": 0.20171457387796268,
      "grad_norm": 0.80078125,
      "learning_rate": 9.95e-05,
      "loss": 7.7332,
      "mean_token_accuracy": 0.04960624761879444,
      "num_tokens": 443940.0,
      "step": 200
    },
    {
      "entropy": 8.258826065063477,
      "epoch": 0.20675743822491174,
      "grad_norm": 0.58984375,
      "learning_rate": 0.000102,
      "loss": 7.6401,
      "mean_token_accuracy": 0.048060222342610356,
      "num_tokens": 455016.0,
      "step": 205
    },
    {
      "entropy": 8.232400035858154,
      "epoch": 0.2118003025718608,
      "grad_norm": 0.68359375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.6756,
      "mean_token_accuracy": 0.05087458901107311,
      "num_tokens": 466380.0,
      "step": 210
    },
    {
      "entropy": 8.21111011505127,
      "epoch": 0.2168431669188099,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000107,
      "loss": 7.5595,
      "mean_token_accuracy": 0.049945533461868766,
      "num_tokens": 476915.0,
      "step": 215
    },
    {
      "entropy": 8.243093967437744,
      "epoch": 0.22188603126575895,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001095,
      "loss": 7.6386,
      "mean_token_accuracy": 0.057695179060101506,
      "num_tokens": 488924.0,
      "step": 220
    },
    {
      "entropy": 8.072712755203247,
      "epoch": 0.22692889561270801,
      "grad_norm": 1.109375,
      "learning_rate": 0.000112,
      "loss": 7.5785,
      "mean_token_accuracy": 0.05549042522907257,
      "num_tokens": 498965.0,
      "step": 225
    },
    {
      "entropy": 8.176321601867675,
      "epoch": 0.23197175995965708,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001145,
      "loss": 7.5965,
      "mean_token_accuracy": 0.056973418965935704,
      "num_tokens": 509755.0,
      "step": 230
    },
    {
      "entropy": 8.122655010223388,
      "epoch": 0.23701462430660616,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.5713,
      "mean_token_accuracy": 0.05634952038526535,
      "num_tokens": 521221.0,
      "step": 235
    },
    {
      "entropy": 8.139823246002198,
      "epoch": 0.24205748865355523,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.6557,
      "mean_token_accuracy": 0.054410668835043904,
      "num_tokens": 533595.0,
      "step": 240
    },
    {
      "entropy": 8.137359714508056,
      "epoch": 0.24710035300050429,
      "grad_norm": 1.03125,
      "learning_rate": 0.000122,
      "loss": 7.5856,
      "mean_token_accuracy": 0.05684904344379902,
      "num_tokens": 543672.0,
      "step": 245
    },
    {
      "entropy": 8.149091720581055,
      "epoch": 0.2521432173474534,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0001245,
      "loss": 7.4872,
      "mean_token_accuracy": 0.06253227479755878,
      "num_tokens": 554906.0,
      "step": 250
    },
    {
      "entropy": 8.084903669357299,
      "epoch": 0.25718608169440244,
      "grad_norm": 0.765625,
      "learning_rate": 0.000127,
      "loss": 7.5417,
      "mean_token_accuracy": 0.06494161449372768,
      "num_tokens": 566419.0,
      "step": 255
    },
    {
      "entropy": 8.07434377670288,
      "epoch": 0.2622289460413515,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001295,
      "loss": 7.5279,
      "mean_token_accuracy": 0.062463199719786644,
      "num_tokens": 578848.0,
      "step": 260
    },
    {
      "entropy": 8.002639818191529,
      "epoch": 0.26727181038830056,
      "grad_norm": 0.9375,
      "learning_rate": 0.000132,
      "loss": 7.4524,
      "mean_token_accuracy": 0.06691358126699924,
      "num_tokens": 590601.0,
      "step": 265
    },
    {
      "entropy": 8.033667945861817,
      "epoch": 0.2723146747352496,
      "grad_norm": 1.25,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.4989,
      "mean_token_accuracy": 0.06721620820462704,
      "num_tokens": 602021.0,
      "step": 270
    },
    {
      "entropy": 7.998079681396485,
      "epoch": 0.2773575390821987,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.5903,
      "mean_token_accuracy": 0.06254490688443184,
      "num_tokens": 613966.0,
      "step": 275
    },
    {
      "entropy": 8.071601390838623,
      "epoch": 0.28240040342914774,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001395,
      "loss": 7.4139,
      "mean_token_accuracy": 0.06817254088819028,
      "num_tokens": 624851.0,
      "step": 280
    },
    {
      "entropy": 7.997099876403809,
      "epoch": 0.2874432677760968,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.4835,
      "mean_token_accuracy": 0.0678780272603035,
      "num_tokens": 635580.0,
      "step": 285
    },
    {
      "entropy": 7.94349856376648,
      "epoch": 0.2924861321230459,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0001445,
      "loss": 7.5534,
      "mean_token_accuracy": 0.06504139304161072,
      "num_tokens": 646927.0,
      "step": 290
    },
    {
      "entropy": 8.021345949172973,
      "epoch": 0.297528996469995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000147,
      "loss": 7.4053,
      "mean_token_accuracy": 0.0697047770023346,
      "num_tokens": 658116.0,
      "step": 295
    },
    {
      "entropy": 7.9564056396484375,
      "epoch": 0.30257186081694404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001495,
      "loss": 7.3887,
      "mean_token_accuracy": 0.06593849398195743,
      "num_tokens": 669054.0,
      "step": 300
    },
    {
      "entropy": 7.9183282375335695,
      "epoch": 0.3076147251638931,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000152,
      "loss": 7.4247,
      "mean_token_accuracy": 0.07052628844976425,
      "num_tokens": 680439.0,
      "step": 305
    },
    {
      "entropy": 7.927852010726928,
      "epoch": 0.31265758951084216,
      "grad_norm": 0.984375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.3485,
      "mean_token_accuracy": 0.0716003067791462,
      "num_tokens": 690909.0,
      "step": 310
    },
    {
      "entropy": 7.843877267837525,
      "epoch": 0.3177004538577912,
      "grad_norm": 1.109375,
      "learning_rate": 0.000157,
      "loss": 7.3335,
      "mean_token_accuracy": 0.0685461949557066,
      "num_tokens": 701744.0,
      "step": 315
    },
    {
      "entropy": 7.78822283744812,
      "epoch": 0.3227433182047403,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0001595,
      "loss": 7.3599,
      "mean_token_accuracy": 0.07047874368727207,
      "num_tokens": 712268.0,
      "step": 320
    },
    {
      "entropy": 7.879432201385498,
      "epoch": 0.32778618255168934,
      "grad_norm": 1.15625,
      "learning_rate": 0.000162,
      "loss": 7.3937,
      "mean_token_accuracy": 0.06803249828517437,
      "num_tokens": 722978.0,
      "step": 325
    },
    {
      "entropy": 7.837024545669555,
      "epoch": 0.3328290468986384,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.3977,
      "mean_token_accuracy": 0.06485099494457244,
      "num_tokens": 734278.0,
      "step": 330
    },
    {
      "entropy": 7.870750522613525,
      "epoch": 0.3378719112455875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3428,
      "mean_token_accuracy": 0.06625398620963097,
      "num_tokens": 746109.0,
      "step": 335
    },
    {
      "entropy": 7.815386438369751,
      "epoch": 0.3429147755925366,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2712,
      "mean_token_accuracy": 0.0703127596527338,
      "num_tokens": 757519.0,
      "step": 340
    },
    {
      "entropy": 7.838667201995849,
      "epoch": 0.34795763993948564,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2327,
      "mean_token_accuracy": 0.06976432576775551,
      "num_tokens": 767665.0,
      "step": 345
    },
    {
      "entropy": 7.770452690124512,
      "epoch": 0.3530005042864347,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.2605,
      "mean_token_accuracy": 0.07452492453157902,
      "num_tokens": 778880.0,
      "step": 350
    },
    {
      "entropy": 7.671797084808349,
      "epoch": 0.35804336863338376,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000177,
      "loss": 7.2654,
      "mean_token_accuracy": 0.07736610174179077,
      "num_tokens": 790360.0,
      "step": 355
    },
    {
      "entropy": 7.728646850585937,
      "epoch": 0.3630862329803328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001795,
      "loss": 7.2348,
      "mean_token_accuracy": 0.07338146939873695,
      "num_tokens": 800889.0,
      "step": 360
    },
    {
      "entropy": 7.675601196289063,
      "epoch": 0.3681290973272819,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000182,
      "loss": 7.2333,
      "mean_token_accuracy": 0.07188612334430218,
      "num_tokens": 811414.0,
      "step": 365
    },
    {
      "entropy": 7.728438234329223,
      "epoch": 0.37317196167423095,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001845,
      "loss": 7.2265,
      "mean_token_accuracy": 0.07272094301879406,
      "num_tokens": 822822.0,
      "step": 370
    },
    {
      "entropy": 7.590001153945923,
      "epoch": 0.37821482602118,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000187,
      "loss": 7.1402,
      "mean_token_accuracy": 0.07824505940079689,
      "num_tokens": 833145.0,
      "step": 375
    },
    {
      "entropy": 7.747368717193604,
      "epoch": 0.3832576903681291,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001895,
      "loss": 7.2475,
      "mean_token_accuracy": 0.0740690290927887,
      "num_tokens": 844015.0,
      "step": 380
    },
    {
      "entropy": 7.641835784912109,
      "epoch": 0.3883005547150782,
      "grad_norm": 0.7578125,
      "learning_rate": 0.000192,
      "loss": 7.1787,
      "mean_token_accuracy": 0.07636387497186661,
      "num_tokens": 855806.0,
      "step": 385
    },
    {
      "entropy": 7.64897780418396,
      "epoch": 0.39334341906202724,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001945,
      "loss": 7.098,
      "mean_token_accuracy": 0.07710095196962356,
      "num_tokens": 865750.0,
      "step": 390
    },
    {
      "entropy": 7.62802677154541,
      "epoch": 0.3983862834089763,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.2246,
      "mean_token_accuracy": 0.07233364507555962,
      "num_tokens": 877630.0,
      "step": 395
    },
    {
      "entropy": 7.64155797958374,
      "epoch": 0.40342914775592537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1193,
      "mean_token_accuracy": 0.07785095870494843,
      "num_tokens": 888667.0,
      "step": 400
    },
    {
      "entropy": 7.625726747512817,
      "epoch": 0.4084720121028744,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000202,
      "loss": 7.197,
      "mean_token_accuracy": 0.07570679485797882,
      "num_tokens": 900989.0,
      "step": 405
    },
    {
      "entropy": 7.635288333892822,
      "epoch": 0.4135148764498235,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.0775,
      "mean_token_accuracy": 0.07197484150528907,
      "num_tokens": 911479.0,
      "step": 410
    },
    {
      "entropy": 7.58501181602478,
      "epoch": 0.41855774079677255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000207,
      "loss": 7.1204,
      "mean_token_accuracy": 0.07360159009695053,
      "num_tokens": 922434.0,
      "step": 415
    },
    {
      "entropy": 7.5460936546325685,
      "epoch": 0.4236006051437216,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002095,
      "loss": 7.1144,
      "mean_token_accuracy": 0.07529124468564988,
      "num_tokens": 933405.0,
      "step": 420
    },
    {
      "entropy": 7.6064252853393555,
      "epoch": 0.4286434694906707,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000212,
      "loss": 7.0975,
      "mean_token_accuracy": 0.07624291330575943,
      "num_tokens": 943567.0,
      "step": 425
    },
    {
      "entropy": 7.480703639984131,
      "epoch": 0.4336863338376198,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002145,
      "loss": 7.136,
      "mean_token_accuracy": 0.07384473346173763,
      "num_tokens": 955123.0,
      "step": 430
    },
    {
      "entropy": 7.631748962402344,
      "epoch": 0.43872919818456885,
      "grad_norm": 1.5,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1533,
      "mean_token_accuracy": 0.07767361216247082,
      "num_tokens": 965565.0,
      "step": 435
    },
    {
      "entropy": 7.526004362106323,
      "epoch": 0.4437720625315179,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002195,
      "loss": 7.1039,
      "mean_token_accuracy": 0.07384410724043847,
      "num_tokens": 977101.0,
      "step": 440
    },
    {
      "entropy": 7.576969480514526,
      "epoch": 0.44881492687846697,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000222,
      "loss": 7.1376,
      "mean_token_accuracy": 0.07549539841711521,
      "num_tokens": 987445.0,
      "step": 445
    },
    {
      "entropy": 7.526590204238891,
      "epoch": 0.45385779122541603,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002245,
      "loss": 7.0196,
      "mean_token_accuracy": 0.08274288028478623,
      "num_tokens": 999379.0,
      "step": 450
    },
    {
      "entropy": 7.471691417694092,
      "epoch": 0.4589006555723651,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.968,
      "mean_token_accuracy": 0.08329149559140206,
      "num_tokens": 1010744.0,
      "step": 455
    },
    {
      "entropy": 7.54520697593689,
      "epoch": 0.46394351991931415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.0416,
      "mean_token_accuracy": 0.07827929966151714,
      "num_tokens": 1021808.0,
      "step": 460
    },
    {
      "entropy": 7.477932929992676,
      "epoch": 0.4689863842662632,
      "grad_norm": 1.015625,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.0811,
      "mean_token_accuracy": 0.07999052368104458,
      "num_tokens": 1033694.0,
      "step": 465
    },
    {
      "entropy": 7.520880126953125,
      "epoch": 0.47402924861321233,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.0456,
      "mean_token_accuracy": 0.07764143832027912,
      "num_tokens": 1044690.0,
      "step": 470
    },
    {
      "entropy": 7.542422342300415,
      "epoch": 0.4790721129601614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000237,
      "loss": 7.0482,
      "mean_token_accuracy": 0.07791347205638885,
      "num_tokens": 1055557.0,
      "step": 475
    },
    {
      "entropy": 7.409520292282105,
      "epoch": 0.48411497730711045,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002395,
      "loss": 6.9957,
      "mean_token_accuracy": 0.08100500479340553,
      "num_tokens": 1067304.0,
      "step": 480
    },
    {
      "entropy": 7.459061002731323,
      "epoch": 0.4891578416540595,
      "grad_norm": 1.0625,
      "learning_rate": 0.000242,
      "loss": 7.021,
      "mean_token_accuracy": 0.0783581517636776,
      "num_tokens": 1078955.0,
      "step": 485
    },
    {
      "entropy": 7.449764680862427,
      "epoch": 0.49420070600100857,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002445,
      "loss": 7.0517,
      "mean_token_accuracy": 0.08230436109006405,
      "num_tokens": 1091405.0,
      "step": 490
    },
    {
      "entropy": 7.463499355316162,
      "epoch": 0.49924357034795763,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000247,
      "loss": 7.0248,
      "mean_token_accuracy": 0.08033888339996338,
      "num_tokens": 1103227.0,
      "step": 495
    },
    {
      "entropy": 7.433458662033081,
      "epoch": 0.5042864346949067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002495,
      "loss": 6.9657,
      "mean_token_accuracy": 0.0861723892390728,
      "num_tokens": 1114485.0,
      "step": 500
    },
    {
      "epoch": 0.5042864346949067,
      "eval_entropy": 7.395277802270788,
      "eval_loss": 7.067502975463867,
      "eval_mean_token_accuracy": 0.08697184946852482,
      "eval_num_tokens": 1114485.0,
      "eval_runtime": 2.1239,
      "eval_samples_per_second": 1477.45,
      "eval_steps_per_second": 185.034,
      "step": 500
    },
    {
      "entropy": 7.405847024917603,
      "epoch": 0.5093292990418558,
      "grad_norm": 1.0625,
      "learning_rate": 0.000252,
      "loss": 6.984,
      "mean_token_accuracy": 0.08936863020062447,
      "num_tokens": 1126789.0,
      "step": 505
    },
    {
      "entropy": 7.460533142089844,
      "epoch": 0.5143721633888049,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002545,
      "loss": 6.966,
      "mean_token_accuracy": 0.08199240490794182,
      "num_tokens": 1138192.0,
      "step": 510
    },
    {
      "entropy": 7.3393408298492435,
      "epoch": 0.5194150277357539,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000257,
      "loss": 6.9098,
      "mean_token_accuracy": 0.08162495531141759,
      "num_tokens": 1148526.0,
      "step": 515
    },
    {
      "entropy": 7.458217191696167,
      "epoch": 0.524457892082703,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002595,
      "loss": 7.0153,
      "mean_token_accuracy": 0.07633640989661217,
      "num_tokens": 1159194.0,
      "step": 520
    },
    {
      "entropy": 7.33443512916565,
      "epoch": 0.529500756429652,
      "grad_norm": 1.203125,
      "learning_rate": 0.000262,
      "loss": 6.906,
      "mean_token_accuracy": 0.0897760234773159,
      "num_tokens": 1170601.0,
      "step": 525
    },
    {
      "entropy": 7.390394878387451,
      "epoch": 0.5345436207766011,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.9091,
      "mean_token_accuracy": 0.08475598022341728,
      "num_tokens": 1182441.0,
      "step": 530
    },
    {
      "entropy": 7.325603246688843,
      "epoch": 0.5395864851235502,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8008,
      "mean_token_accuracy": 0.09372416436672211,
      "num_tokens": 1192530.0,
      "step": 535
    },
    {
      "entropy": 7.284105587005615,
      "epoch": 0.5446293494704992,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.9239,
      "mean_token_accuracy": 0.08224731758236885,
      "num_tokens": 1203831.0,
      "step": 540
    },
    {
      "entropy": 7.360192966461182,
      "epoch": 0.5496722138174484,
      "grad_norm": 1.25,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8601,
      "mean_token_accuracy": 0.08579591512680054,
      "num_tokens": 1214164.0,
      "step": 545
    },
    {
      "entropy": 7.267377424240112,
      "epoch": 0.5547150781643974,
      "grad_norm": 1.0,
      "learning_rate": 0.0002745,
      "loss": 6.8831,
      "mean_token_accuracy": 0.08585480228066444,
      "num_tokens": 1226932.0,
      "step": 550
    },
    {
      "entropy": 7.379234743118286,
      "epoch": 0.5597579425113465,
      "grad_norm": 1.15625,
      "learning_rate": 0.000277,
      "loss": 6.942,
      "mean_token_accuracy": 0.0817857712507248,
      "num_tokens": 1237212.0,
      "step": 555
    },
    {
      "entropy": 7.295353126525879,
      "epoch": 0.5648008068582955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002795,
      "loss": 6.8459,
      "mean_token_accuracy": 0.08828911706805229,
      "num_tokens": 1249340.0,
      "step": 560
    },
    {
      "entropy": 7.293833637237549,
      "epoch": 0.5698436712052446,
      "grad_norm": 1.65625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.8602,
      "mean_token_accuracy": 0.0935377262532711,
      "num_tokens": 1260778.0,
      "step": 565
    },
    {
      "entropy": 7.2916364669799805,
      "epoch": 0.5748865355521936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002845,
      "loss": 6.8675,
      "mean_token_accuracy": 0.0834194079041481,
      "num_tokens": 1272216.0,
      "step": 570
    },
    {
      "entropy": 7.310083103179932,
      "epoch": 0.5799293998991427,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000287,
      "loss": 6.9369,
      "mean_token_accuracy": 0.08788081333041191,
      "num_tokens": 1283558.0,
      "step": 575
    },
    {
      "entropy": 7.404259157180786,
      "epoch": 0.5849722642460918,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002895,
      "loss": 6.9825,
      "mean_token_accuracy": 0.08486812300980091,
      "num_tokens": 1294638.0,
      "step": 580
    },
    {
      "entropy": 7.435716342926026,
      "epoch": 0.5900151285930408,
      "grad_norm": 1.15625,
      "learning_rate": 0.000292,
      "loss": 6.8912,
      "mean_token_accuracy": 0.087406737357378,
      "num_tokens": 1304697.0,
      "step": 585
    },
    {
      "entropy": 7.253348064422608,
      "epoch": 0.59505799293999,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002945,
      "loss": 6.8609,
      "mean_token_accuracy": 0.08688995763659477,
      "num_tokens": 1316618.0,
      "step": 590
    },
    {
      "entropy": 7.336230707168579,
      "epoch": 0.600100857286939,
      "grad_norm": 1.140625,
      "learning_rate": 0.000297,
      "loss": 6.9537,
      "mean_token_accuracy": 0.08084713369607925,
      "num_tokens": 1328134.0,
      "step": 595
    },
    {
      "entropy": 7.371943950653076,
      "epoch": 0.6051437216338881,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002995,
      "loss": 6.8797,
      "mean_token_accuracy": 0.08257691636681556,
      "num_tokens": 1339101.0,
      "step": 600
    },
    {
      "entropy": 7.2372519969940186,
      "epoch": 0.6101865859808371,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000302,
      "loss": 6.7574,
      "mean_token_accuracy": 0.09110561460256576,
      "num_tokens": 1349181.0,
      "step": 605
    },
    {
      "entropy": 7.217679405212403,
      "epoch": 0.6152294503277862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045,
      "loss": 6.8774,
      "mean_token_accuracy": 0.08801197558641434,
      "num_tokens": 1361443.0,
      "step": 610
    },
    {
      "entropy": 7.238507413864136,
      "epoch": 0.6202723146747352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000307,
      "loss": 6.8177,
      "mean_token_accuracy": 0.09036832302808762,
      "num_tokens": 1372807.0,
      "step": 615
    },
    {
      "entropy": 7.285562324523926,
      "epoch": 0.6253151790216843,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003095,
      "loss": 6.8942,
      "mean_token_accuracy": 0.08551665917038917,
      "num_tokens": 1385509.0,
      "step": 620
    },
    {
      "entropy": 7.294302272796631,
      "epoch": 0.6303580433686334,
      "grad_norm": 1.171875,
      "learning_rate": 0.000312,
      "loss": 6.7994,
      "mean_token_accuracy": 0.09008246585726738,
      "num_tokens": 1396035.0,
      "step": 625
    },
    {
      "entropy": 7.160209274291992,
      "epoch": 0.6354009077155824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003145,
      "loss": 6.7335,
      "mean_token_accuracy": 0.0913459986448288,
      "num_tokens": 1406839.0,
      "step": 630
    },
    {
      "entropy": 7.1683602809906,
      "epoch": 0.6404437720625316,
      "grad_norm": 1.265625,
      "learning_rate": 0.000317,
      "loss": 6.7121,
      "mean_token_accuracy": 0.09201777651906014,
      "num_tokens": 1417708.0,
      "step": 635
    },
    {
      "entropy": 7.210118246078491,
      "epoch": 0.6454866364094806,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003195,
      "loss": 6.8383,
      "mean_token_accuracy": 0.08575858026742936,
      "num_tokens": 1428985.0,
      "step": 640
    },
    {
      "entropy": 7.197644662857056,
      "epoch": 0.6505295007564297,
      "grad_norm": 1.3125,
      "learning_rate": 0.000322,
      "loss": 6.8489,
      "mean_token_accuracy": 0.09253060892224312,
      "num_tokens": 1440232.0,
      "step": 645
    },
    {
      "entropy": 7.127893161773682,
      "epoch": 0.6555723651033787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6128,
      "mean_token_accuracy": 0.09752296134829522,
      "num_tokens": 1451735.0,
      "step": 650
    },
    {
      "entropy": 7.12758059501648,
      "epoch": 0.6606152294503278,
      "grad_norm": 1.125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7665,
      "mean_token_accuracy": 0.0905262552201748,
      "num_tokens": 1463477.0,
      "step": 655
    },
    {
      "entropy": 7.16529951095581,
      "epoch": 0.6656580937972768,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.7643,
      "mean_token_accuracy": 0.08742906525731087,
      "num_tokens": 1473802.0,
      "step": 660
    },
    {
      "entropy": 7.178797626495362,
      "epoch": 0.6707009581442259,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7625,
      "mean_token_accuracy": 0.09097197279334068,
      "num_tokens": 1485519.0,
      "step": 665
    },
    {
      "entropy": 7.134563350677491,
      "epoch": 0.675743822491175,
      "grad_norm": 0.875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.7571,
      "mean_token_accuracy": 0.09082425311207772,
      "num_tokens": 1496957.0,
      "step": 670
    },
    {
      "entropy": 7.125902271270752,
      "epoch": 0.680786686838124,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000337,
      "loss": 6.7513,
      "mean_token_accuracy": 0.09530752003192902,
      "num_tokens": 1508336.0,
      "step": 675
    },
    {
      "entropy": 7.134354066848755,
      "epoch": 0.6858295511850732,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003395,
      "loss": 6.7368,
      "mean_token_accuracy": 0.09136363193392753,
      "num_tokens": 1520016.0,
      "step": 680
    },
    {
      "entropy": 7.047116708755493,
      "epoch": 0.6908724155320222,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000342,
      "loss": 6.5969,
      "mean_token_accuracy": 0.10183367505669594,
      "num_tokens": 1531877.0,
      "step": 685
    },
    {
      "entropy": 6.992479372024536,
      "epoch": 0.6959152798789713,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5929,
      "mean_token_accuracy": 0.09865794181823731,
      "num_tokens": 1541553.0,
      "step": 690
    },
    {
      "entropy": 7.179808568954468,
      "epoch": 0.7009581442259203,
      "grad_norm": 1.140625,
      "learning_rate": 0.000347,
      "loss": 6.7486,
      "mean_token_accuracy": 0.09588267058134078,
      "num_tokens": 1552491.0,
      "step": 695
    },
    {
      "entropy": 6.9392517566680905,
      "epoch": 0.7060010085728694,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003495,
      "loss": 6.6438,
      "mean_token_accuracy": 0.09699270352721215,
      "num_tokens": 1563695.0,
      "step": 700
    },
    {
      "entropy": 7.107646083831787,
      "epoch": 0.7110438729198184,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000352,
      "loss": 6.6743,
      "mean_token_accuracy": 0.0914404772222042,
      "num_tokens": 1573467.0,
      "step": 705
    },
    {
      "entropy": 6.988015794754029,
      "epoch": 0.7160867372667675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003545,
      "loss": 6.6471,
      "mean_token_accuracy": 0.10227895081043244,
      "num_tokens": 1584367.0,
      "step": 710
    },
    {
      "entropy": 7.015262031555176,
      "epoch": 0.7211296016137166,
      "grad_norm": 1.140625,
      "learning_rate": 0.000357,
      "loss": 6.5912,
      "mean_token_accuracy": 0.10431398674845696,
      "num_tokens": 1595301.0,
      "step": 715
    },
    {
      "entropy": 7.049318599700928,
      "epoch": 0.7261724659606656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003595,
      "loss": 6.6489,
      "mean_token_accuracy": 0.09553751796483993,
      "num_tokens": 1605916.0,
      "step": 720
    },
    {
      "entropy": 7.039408111572266,
      "epoch": 0.7312153303076148,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000362,
      "loss": 6.6366,
      "mean_token_accuracy": 0.09584024250507354,
      "num_tokens": 1616455.0,
      "step": 725
    },
    {
      "entropy": 6.976061582565308,
      "epoch": 0.7362581946545638,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003645,
      "loss": 6.6249,
      "mean_token_accuracy": 0.10145259872078896,
      "num_tokens": 1627396.0,
      "step": 730
    },
    {
      "entropy": 7.117921972274781,
      "epoch": 0.7413010590015129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000367,
      "loss": 6.7499,
      "mean_token_accuracy": 0.09127496853470803,
      "num_tokens": 1639073.0,
      "step": 735
    },
    {
      "entropy": 6.956200408935547,
      "epoch": 0.7463439233484619,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695,
      "loss": 6.6271,
      "mean_token_accuracy": 0.09801545068621635,
      "num_tokens": 1649787.0,
      "step": 740
    },
    {
      "entropy": 6.9809126377105715,
      "epoch": 0.751386787695411,
      "grad_norm": 1.109375,
      "learning_rate": 0.000372,
      "loss": 6.5655,
      "mean_token_accuracy": 0.09974676370620728,
      "num_tokens": 1658883.0,
      "step": 745
    },
    {
      "entropy": 6.860607004165649,
      "epoch": 0.75642965204236,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003745,
      "loss": 6.5439,
      "mean_token_accuracy": 0.1046278566122055,
      "num_tokens": 1669099.0,
      "step": 750
    },
    {
      "entropy": 6.945769739151001,
      "epoch": 0.7614725163893091,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000377,
      "loss": 6.5476,
      "mean_token_accuracy": 0.09769671782851219,
      "num_tokens": 1680227.0,
      "step": 755
    },
    {
      "entropy": 6.982058095932007,
      "epoch": 0.7665153807362582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003795,
      "loss": 6.5253,
      "mean_token_accuracy": 0.10293467938899994,
      "num_tokens": 1689901.0,
      "step": 760
    },
    {
      "entropy": 6.859155321121216,
      "epoch": 0.7715582450832073,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000382,
      "loss": 6.5986,
      "mean_token_accuracy": 0.09999214485287666,
      "num_tokens": 1700498.0,
      "step": 765
    },
    {
      "entropy": 6.901654386520386,
      "epoch": 0.7766011094301564,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003845,
      "loss": 6.5141,
      "mean_token_accuracy": 0.10296274051070213,
      "num_tokens": 1710715.0,
      "step": 770
    },
    {
      "entropy": 6.941842555999756,
      "epoch": 0.7816439737771054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.6341,
      "mean_token_accuracy": 0.09950209185481071,
      "num_tokens": 1723858.0,
      "step": 775
    },
    {
      "entropy": 6.89209246635437,
      "epoch": 0.7866868381240545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5935,
      "mean_token_accuracy": 0.09372718334197998,
      "num_tokens": 1734678.0,
      "step": 780
    },
    {
      "entropy": 6.904127931594848,
      "epoch": 0.7917297024710035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5056,
      "mean_token_accuracy": 0.1047722913324833,
      "num_tokens": 1743869.0,
      "step": 785
    },
    {
      "entropy": 6.984526538848877,
      "epoch": 0.7967725668179526,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.578,
      "mean_token_accuracy": 0.09974783658981323,
      "num_tokens": 1755266.0,
      "step": 790
    },
    {
      "entropy": 6.783455896377563,
      "epoch": 0.8018154311649016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5364,
      "mean_token_accuracy": 0.09974951297044754,
      "num_tokens": 1766089.0,
      "step": 795
    },
    {
      "entropy": 6.944797992706299,
      "epoch": 0.8068582955118507,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003995,
      "loss": 6.5431,
      "mean_token_accuracy": 0.10037100091576576,
      "num_tokens": 1776057.0,
      "step": 800
    },
    {
      "entropy": 6.836930894851685,
      "epoch": 0.8119011598587998,
      "grad_norm": 1.1875,
      "learning_rate": 0.000402,
      "loss": 6.5737,
      "mean_token_accuracy": 0.09728048965334893,
      "num_tokens": 1786729.0,
      "step": 805
    },
    {
      "entropy": 6.9329804420471195,
      "epoch": 0.8169440242057489,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004045,
      "loss": 6.508,
      "mean_token_accuracy": 0.10167987123131753,
      "num_tokens": 1797708.0,
      "step": 810
    },
    {
      "entropy": 6.859876441955566,
      "epoch": 0.821986888552698,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5453,
      "mean_token_accuracy": 0.10706769153475762,
      "num_tokens": 1807326.0,
      "step": 815
    },
    {
      "entropy": 6.783277606964111,
      "epoch": 0.827029752899647,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004095,
      "loss": 6.4844,
      "mean_token_accuracy": 0.1034775085747242,
      "num_tokens": 1817266.0,
      "step": 820
    },
    {
      "entropy": 6.894398736953735,
      "epoch": 0.8320726172465961,
      "grad_norm": 1.09375,
      "learning_rate": 0.000412,
      "loss": 6.5,
      "mean_token_accuracy": 0.10923754200339317,
      "num_tokens": 1827840.0,
      "step": 825
    },
    {
      "entropy": 6.709803342819214,
      "epoch": 0.8371154815935451,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004145,
      "loss": 6.4748,
      "mean_token_accuracy": 0.1036040298640728,
      "num_tokens": 1838158.0,
      "step": 830
    },
    {
      "entropy": 6.877896070480347,
      "epoch": 0.8421583459404942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000417,
      "loss": 6.5591,
      "mean_token_accuracy": 0.09829444661736489,
      "num_tokens": 1848317.0,
      "step": 835
    },
    {
      "entropy": 6.768724632263184,
      "epoch": 0.8472012102874432,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004195,
      "loss": 6.4806,
      "mean_token_accuracy": 0.10388926193118095,
      "num_tokens": 1858751.0,
      "step": 840
    },
    {
      "entropy": 6.846293878555298,
      "epoch": 0.8522440746343923,
      "grad_norm": 1.03125,
      "learning_rate": 0.000422,
      "loss": 6.493,
      "mean_token_accuracy": 0.10360832959413528,
      "num_tokens": 1870739.0,
      "step": 845
    },
    {
      "entropy": 6.887697744369507,
      "epoch": 0.8572869389813415,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004245,
      "loss": 6.6064,
      "mean_token_accuracy": 0.10239030867815017,
      "num_tokens": 1880995.0,
      "step": 850
    },
    {
      "entropy": 6.8503830909729,
      "epoch": 0.8623298033282905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000427,
      "loss": 6.5576,
      "mean_token_accuracy": 0.1004607506096363,
      "num_tokens": 1892633.0,
      "step": 855
    },
    {
      "entropy": 6.927896118164062,
      "epoch": 0.8673726676752396,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004295,
      "loss": 6.6157,
      "mean_token_accuracy": 0.09754905551671982,
      "num_tokens": 1904802.0,
      "step": 860
    },
    {
      "entropy": 6.896012830734253,
      "epoch": 0.8724155320221886,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432,
      "loss": 6.6147,
      "mean_token_accuracy": 0.1034881442785263,
      "num_tokens": 1915755.0,
      "step": 865
    },
    {
      "entropy": 6.753184366226196,
      "epoch": 0.8774583963691377,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004345,
      "loss": 6.3595,
      "mean_token_accuracy": 0.1010950192809105,
      "num_tokens": 1926098.0,
      "step": 870
    },
    {
      "entropy": 6.809195327758789,
      "epoch": 0.8825012607160867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000437,
      "loss": 6.5846,
      "mean_token_accuracy": 0.1017203226685524,
      "num_tokens": 1936607.0,
      "step": 875
    },
    {
      "entropy": 6.739422082901001,
      "epoch": 0.8875441250630358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004395,
      "loss": 6.4709,
      "mean_token_accuracy": 0.10663464665412903,
      "num_tokens": 1948596.0,
      "step": 880
    },
    {
      "entropy": 6.76585898399353,
      "epoch": 0.8925869894099848,
      "grad_norm": 1.109375,
      "learning_rate": 0.000442,
      "loss": 6.5381,
      "mean_token_accuracy": 0.1049344077706337,
      "num_tokens": 1960322.0,
      "step": 885
    },
    {
      "entropy": 6.844669008255005,
      "epoch": 0.8976298537569339,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004445,
      "loss": 6.4564,
      "mean_token_accuracy": 0.10585802644491196,
      "num_tokens": 1971052.0,
      "step": 890
    },
    {
      "entropy": 6.804664039611817,
      "epoch": 0.902672718103883,
      "grad_norm": 1.125,
      "learning_rate": 0.000447,
      "loss": 6.4861,
      "mean_token_accuracy": 0.09895869046449661,
      "num_tokens": 1980917.0,
      "step": 895
    },
    {
      "entropy": 6.729268407821655,
      "epoch": 0.9077155824508321,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4384,
      "mean_token_accuracy": 0.10832197889685631,
      "num_tokens": 1991035.0,
      "step": 900
    },
    {
      "entropy": 6.7104339599609375,
      "epoch": 0.9127584467977812,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.458,
      "mean_token_accuracy": 0.10516942515969277,
      "num_tokens": 2000587.0,
      "step": 905
    },
    {
      "entropy": 6.7780543804168705,
      "epoch": 0.9178013111447302,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4273,
      "mean_token_accuracy": 0.10677500739693642,
      "num_tokens": 2011087.0,
      "step": 910
    },
    {
      "entropy": 6.705195903778076,
      "epoch": 0.9228441754916793,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4673,
      "mean_token_accuracy": 0.10288073197007179,
      "num_tokens": 2023472.0,
      "step": 915
    },
    {
      "entropy": 6.75528621673584,
      "epoch": 0.9278870398386283,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.454,
      "mean_token_accuracy": 0.10575027167797088,
      "num_tokens": 2033783.0,
      "step": 920
    },
    {
      "entropy": 6.657633304595947,
      "epoch": 0.9329299041855774,
      "grad_norm": 1.046875,
      "learning_rate": 0.000462,
      "loss": 6.4729,
      "mean_token_accuracy": 0.10640397965908051,
      "num_tokens": 2045882.0,
      "step": 925
    },
    {
      "entropy": 6.779510593414306,
      "epoch": 0.9379727685325264,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004645,
      "loss": 6.4781,
      "mean_token_accuracy": 0.10185832753777505,
      "num_tokens": 2056549.0,
      "step": 930
    },
    {
      "entropy": 6.75550947189331,
      "epoch": 0.9430156328794755,
      "grad_norm": 1.265625,
      "learning_rate": 0.000467,
      "loss": 6.4788,
      "mean_token_accuracy": 0.10527065321803093,
      "num_tokens": 2067540.0,
      "step": 935
    },
    {
      "entropy": 6.6518762588500975,
      "epoch": 0.9480584972264247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004695,
      "loss": 6.3729,
      "mean_token_accuracy": 0.10782607644796371,
      "num_tokens": 2077386.0,
      "step": 940
    },
    {
      "entropy": 6.694978809356689,
      "epoch": 0.9531013615733737,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000472,
      "loss": 6.3588,
      "mean_token_accuracy": 0.10642436146736145,
      "num_tokens": 2087777.0,
      "step": 945
    },
    {
      "entropy": 6.664656162261963,
      "epoch": 0.9581442259203228,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004745,
      "loss": 6.351,
      "mean_token_accuracy": 0.10316140428185464,
      "num_tokens": 2099071.0,
      "step": 950
    },
    {
      "entropy": 6.4923890113830565,
      "epoch": 0.9631870902672718,
      "grad_norm": 1.140625,
      "learning_rate": 0.000477,
      "loss": 6.4144,
      "mean_token_accuracy": 0.10309024602174759,
      "num_tokens": 2109168.0,
      "step": 955
    },
    {
      "entropy": 6.833002996444702,
      "epoch": 0.9682299546142209,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795,
      "loss": 6.5138,
      "mean_token_accuracy": 0.1049713984131813,
      "num_tokens": 2120696.0,
      "step": 960
    },
    {
      "entropy": 6.668496990203858,
      "epoch": 0.9732728189611699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000482,
      "loss": 6.4429,
      "mean_token_accuracy": 0.10518230870366096,
      "num_tokens": 2132427.0,
      "step": 965
    },
    {
      "entropy": 6.701524496078491,
      "epoch": 0.978315683308119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845,
      "loss": 6.4353,
      "mean_token_accuracy": 0.10702200308442116,
      "num_tokens": 2143120.0,
      "step": 970
    },
    {
      "entropy": 6.689940118789673,
      "epoch": 0.983358547655068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487,
      "loss": 6.4142,
      "mean_token_accuracy": 0.10534614622592926,
      "num_tokens": 2154154.0,
      "step": 975
    },
    {
      "entropy": 6.696073293685913,
      "epoch": 0.9884014120020171,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895,
      "loss": 6.5324,
      "mean_token_accuracy": 0.0996658444404602,
      "num_tokens": 2167000.0,
      "step": 980
    },
    {
      "entropy": 6.711336994171143,
      "epoch": 0.9934442763489663,
      "grad_norm": 1.15625,
      "learning_rate": 0.000492,
      "loss": 6.4649,
      "mean_token_accuracy": 0.10657059699296952,
      "num_tokens": 2178100.0,
      "step": 985
    },
    {
      "entropy": 6.600247192382812,
      "epoch": 0.9984871406959153,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004945,
      "loss": 6.4187,
      "mean_token_accuracy": 0.10585904195904731,
      "num_tokens": 2189063.0,
      "step": 990
    },
    {
      "entropy": 6.7853367063734265,
      "epoch": 1.0030257186081695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497,
      "loss": 6.3816,
      "mean_token_accuracy": 0.10893791500065061,
      "num_tokens": 2198302.0,
      "step": 995
    },
    {
      "entropy": 6.630726861953735,
      "epoch": 1.0080685829551186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995,
      "loss": 6.3312,
      "mean_token_accuracy": 0.1090955302119255,
      "num_tokens": 2209648.0,
      "step": 1000
    },
    {
      "epoch": 1.0080685829551186,
      "eval_entropy": 6.540408185420145,
      "eval_loss": 6.461416721343994,
      "eval_mean_token_accuracy": 0.11113336287040747,
      "eval_num_tokens": 2209648.0,
      "eval_runtime": 2.0341,
      "eval_samples_per_second": 1542.682,
      "eval_steps_per_second": 193.204,
      "step": 1000
    },
    {
      "entropy": 6.596870613098145,
      "epoch": 1.0131114473020675,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999997762223182,
      "loss": 6.2431,
      "mean_token_accuracy": 0.11784344986081123,
      "num_tokens": 2220806.0,
      "step": 1005
    },
    {
      "entropy": 6.553163862228393,
      "epoch": 1.0181543116490166,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999988671262489,
      "loss": 6.2489,
      "mean_token_accuracy": 0.11016370132565498,
      "num_tokens": 2231806.0,
      "step": 1010
    },
    {
      "entropy": 6.60176043510437,
      "epoch": 1.0231971759959657,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999972587285105,
      "loss": 6.2911,
      "mean_token_accuracy": 0.11688948720693589,
      "num_tokens": 2243835.0,
      "step": 1015
    },
    {
      "entropy": 6.443943977355957,
      "epoch": 1.0282400403429148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999949510341015,
      "loss": 6.2047,
      "mean_token_accuracy": 0.11431019976735116,
      "num_tokens": 2254722.0,
      "step": 1020
    },
    {
      "entropy": 6.5499800682067875,
      "epoch": 1.033282904689864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999919440501947,
      "loss": 6.1699,
      "mean_token_accuracy": 0.12210036814212799,
      "num_tokens": 2264806.0,
      "step": 1025
    },
    {
      "entropy": 6.655332374572754,
      "epoch": 1.0383257690368128,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999882377861355,
      "loss": 6.3597,
      "mean_token_accuracy": 0.10959624648094177,
      "num_tokens": 2277037.0,
      "step": 1030
    },
    {
      "entropy": 6.563287734985352,
      "epoch": 1.043368633383762,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999838322534433,
      "loss": 6.2914,
      "mean_token_accuracy": 0.10317928344011307,
      "num_tokens": 2289829.0,
      "step": 1035
    },
    {
      "entropy": 6.605117607116699,
      "epoch": 1.048411497730711,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999787274658107,
      "loss": 6.1877,
      "mean_token_accuracy": 0.11521424874663352,
      "num_tokens": 2300340.0,
      "step": 1040
    },
    {
      "entropy": 6.508002901077271,
      "epoch": 1.0534543620776602,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999729234391032,
      "loss": 6.2135,
      "mean_token_accuracy": 0.11831212639808655,
      "num_tokens": 2310009.0,
      "step": 1045
    },
    {
      "entropy": 6.558585786819458,
      "epoch": 1.058497226424609,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049996642019136,
      "loss": 6.2327,
      "mean_token_accuracy": 0.11367388069629669,
      "num_tokens": 2319290.0,
      "step": 1050
    },
    {
      "entropy": 6.563412809371949,
      "epoch": 1.0635400907715582,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999592177427933,
      "loss": 6.2095,
      "mean_token_accuracy": 0.11704154834151267,
      "num_tokens": 2329170.0,
      "step": 1055
    },
    {
      "entropy": 6.4528532981872555,
      "epoch": 1.0685829551185073,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999513161157885,
      "loss": 6.2172,
      "mean_token_accuracy": 0.11061301082372665,
      "num_tokens": 2339617.0,
      "step": 1060
    },
    {
      "entropy": 6.584239149093628,
      "epoch": 1.0736258194654564,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999427153349041,
      "loss": 6.2238,
      "mean_token_accuracy": 0.11742238327860832,
      "num_tokens": 2349548.0,
      "step": 1065
    },
    {
      "entropy": 6.460854816436767,
      "epoch": 1.0786686838124055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999334154268714,
      "loss": 6.18,
      "mean_token_accuracy": 0.11255253925919532,
      "num_tokens": 2359116.0,
      "step": 1070
    },
    {
      "entropy": 6.515554475784302,
      "epoch": 1.0837115481593544,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999234164205949,
      "loss": 6.2383,
      "mean_token_accuracy": 0.12132272943854332,
      "num_tokens": 2369412.0,
      "step": 1075
    },
    {
      "entropy": 6.517651605606079,
      "epoch": 1.0887544125063036,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999127183471517,
      "loss": 6.3072,
      "mean_token_accuracy": 0.11325916647911072,
      "num_tokens": 2381276.0,
      "step": 1080
    },
    {
      "entropy": 6.56121187210083,
      "epoch": 1.0937972768532527,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999013212397915,
      "loss": 6.2166,
      "mean_token_accuracy": 0.11995315775275231,
      "num_tokens": 2392135.0,
      "step": 1085
    },
    {
      "entropy": 6.375766229629517,
      "epoch": 1.0988401412002018,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998892251339368,
      "loss": 6.1213,
      "mean_token_accuracy": 0.12119341865181923,
      "num_tokens": 2402957.0,
      "step": 1090
    },
    {
      "entropy": 6.571408224105835,
      "epoch": 1.1038830055471507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998764300671827,
      "loss": 6.251,
      "mean_token_accuracy": 0.12106918767094613,
      "num_tokens": 2413386.0,
      "step": 1095
    },
    {
      "entropy": 6.413351535797119,
      "epoch": 1.1089258698940998,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998629360792965,
      "loss": 6.2446,
      "mean_token_accuracy": 0.11359023898839951,
      "num_tokens": 2424441.0,
      "step": 1100
    },
    {
      "entropy": 6.525660133361816,
      "epoch": 1.113968734241049,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998487432122179,
      "loss": 6.2587,
      "mean_token_accuracy": 0.11170726269483566,
      "num_tokens": 2436144.0,
      "step": 1105
    },
    {
      "entropy": 6.5052775859832765,
      "epoch": 1.119011598587998,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998338515100585,
      "loss": 6.1793,
      "mean_token_accuracy": 0.1222833052277565,
      "num_tokens": 2447393.0,
      "step": 1110
    },
    {
      "entropy": 6.421037435531616,
      "epoch": 1.1240544629349472,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998182610191022,
      "loss": 6.1303,
      "mean_token_accuracy": 0.1208079345524311,
      "num_tokens": 2458632.0,
      "step": 1115
    },
    {
      "entropy": 6.469361782073975,
      "epoch": 1.129097327281896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998019717878047,
      "loss": 6.221,
      "mean_token_accuracy": 0.11303923651576042,
      "num_tokens": 2469331.0,
      "step": 1120
    },
    {
      "entropy": 6.545333385467529,
      "epoch": 1.1341401916288452,
      "grad_norm": 1.125,
      "learning_rate": 0.000499784983866793,
      "loss": 6.2689,
      "mean_token_accuracy": 0.11166509613394737,
      "num_tokens": 2479501.0,
      "step": 1125
    },
    {
      "entropy": 6.431850719451904,
      "epoch": 1.1391830559757943,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997672973088661,
      "loss": 6.1293,
      "mean_token_accuracy": 0.11632360890507698,
      "num_tokens": 2490087.0,
      "step": 1130
    },
    {
      "entropy": 6.484920310974121,
      "epoch": 1.1442259203227434,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997489121689944,
      "loss": 6.2024,
      "mean_token_accuracy": 0.11425360813736915,
      "num_tokens": 2501130.0,
      "step": 1135
    },
    {
      "entropy": 6.501006603240967,
      "epoch": 1.1492687846696923,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997298285043189,
      "loss": 6.2999,
      "mean_token_accuracy": 0.11134214028716087,
      "num_tokens": 2513999.0,
      "step": 1140
    },
    {
      "entropy": 6.474369859695434,
      "epoch": 1.1543116490166414,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997100463741527,
      "loss": 6.2194,
      "mean_token_accuracy": 0.11561294570565224,
      "num_tokens": 2525579.0,
      "step": 1145
    },
    {
      "entropy": 6.4915367603302006,
      "epoch": 1.1593545133635905,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996895658399785,
      "loss": 6.1905,
      "mean_token_accuracy": 0.11502130404114723,
      "num_tokens": 2536091.0,
      "step": 1150
    },
    {
      "entropy": 6.468376922607422,
      "epoch": 1.1643973777105396,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996683869654507,
      "loss": 6.2268,
      "mean_token_accuracy": 0.1178181879222393,
      "num_tokens": 2546073.0,
      "step": 1155
    },
    {
      "entropy": 6.495087146759033,
      "epoch": 1.1694402420574885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996465098163934,
      "loss": 6.1908,
      "mean_token_accuracy": 0.11532010957598686,
      "num_tokens": 2556626.0,
      "step": 1160
    },
    {
      "entropy": 6.426233196258545,
      "epoch": 1.1744831064044376,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996239344608015,
      "loss": 6.1999,
      "mean_token_accuracy": 0.1225629799067974,
      "num_tokens": 2567304.0,
      "step": 1165
    },
    {
      "entropy": 6.414515161514283,
      "epoch": 1.1795259707513868,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996006609688398,
      "loss": 6.0725,
      "mean_token_accuracy": 0.11471373662352562,
      "num_tokens": 2577242.0,
      "step": 1170
    },
    {
      "entropy": 6.41071925163269,
      "epoch": 1.1845688350983359,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995766894128426,
      "loss": 6.1714,
      "mean_token_accuracy": 0.11379559189081193,
      "num_tokens": 2588330.0,
      "step": 1175
    },
    {
      "entropy": 6.4306882381439205,
      "epoch": 1.189611699445285,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995520198673143,
      "loss": 6.1998,
      "mean_token_accuracy": 0.11733074858784676,
      "num_tokens": 2598471.0,
      "step": 1180
    },
    {
      "entropy": 6.489285898208618,
      "epoch": 1.194654563792234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995266524089283,
      "loss": 6.257,
      "mean_token_accuracy": 0.1135383665561676,
      "num_tokens": 2610238.0,
      "step": 1185
    },
    {
      "entropy": 6.389137125015258,
      "epoch": 1.199697428139183,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995005871165271,
      "loss": 6.1335,
      "mean_token_accuracy": 0.12369364947080612,
      "num_tokens": 2619870.0,
      "step": 1190
    },
    {
      "entropy": 6.471128797531128,
      "epoch": 1.2047402924861321,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994738240711225,
      "loss": 6.2621,
      "mean_token_accuracy": 0.11531401872634887,
      "num_tokens": 2631541.0,
      "step": 1195
    },
    {
      "entropy": 6.359742164611816,
      "epoch": 1.2097831568330812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994463633558945,
      "loss": 6.0995,
      "mean_token_accuracy": 0.12339793145656586,
      "num_tokens": 2641896.0,
      "step": 1200
    },
    {
      "entropy": 6.430509996414185,
      "epoch": 1.2148260211800301,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994182050561917,
      "loss": 6.1842,
      "mean_token_accuracy": 0.1165244683623314,
      "num_tokens": 2654208.0,
      "step": 1205
    },
    {
      "entropy": 6.394809579849243,
      "epoch": 1.2198688855269793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993893492595306,
      "loss": 6.1739,
      "mean_token_accuracy": 0.11764412522315978,
      "num_tokens": 2665115.0,
      "step": 1210
    },
    {
      "entropy": 6.455261135101319,
      "epoch": 1.2249117498739284,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993597960555959,
      "loss": 6.2527,
      "mean_token_accuracy": 0.11351834759116172,
      "num_tokens": 2676093.0,
      "step": 1215
    },
    {
      "entropy": 6.452110052108765,
      "epoch": 1.2299546142208775,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993295455362395,
      "loss": 6.197,
      "mean_token_accuracy": 0.12444573938846588,
      "num_tokens": 2687866.0,
      "step": 1220
    },
    {
      "entropy": 6.458801126480102,
      "epoch": 1.2349974785678266,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992985977954808,
      "loss": 6.2066,
      "mean_token_accuracy": 0.11660922318696976,
      "num_tokens": 2699814.0,
      "step": 1225
    },
    {
      "entropy": 6.44323582649231,
      "epoch": 1.2400403429147757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499266952929506,
      "loss": 6.1839,
      "mean_token_accuracy": 0.12365676537156105,
      "num_tokens": 2710384.0,
      "step": 1230
    },
    {
      "entropy": 6.4240641593933105,
      "epoch": 1.2450832072617246,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992346110366682,
      "loss": 6.1554,
      "mean_token_accuracy": 0.11770453900098801,
      "num_tokens": 2721442.0,
      "step": 1235
    },
    {
      "entropy": 6.377686071395874,
      "epoch": 1.2501260716086737,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992015722174865,
      "loss": 6.1657,
      "mean_token_accuracy": 0.11785626113414764,
      "num_tokens": 2733605.0,
      "step": 1240
    },
    {
      "entropy": 6.43232831954956,
      "epoch": 1.2551689359556228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991678365746466,
      "loss": 6.2089,
      "mean_token_accuracy": 0.12633543610572814,
      "num_tokens": 2744853.0,
      "step": 1245
    },
    {
      "entropy": 6.320886278152466,
      "epoch": 1.2602118003025717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991334042129996,
      "loss": 6.1113,
      "mean_token_accuracy": 0.1273192882537842,
      "num_tokens": 2754477.0,
      "step": 1250
    },
    {
      "entropy": 6.442186880111694,
      "epoch": 1.2652546646495209,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990982752395618,
      "loss": 6.1173,
      "mean_token_accuracy": 0.1199355199933052,
      "num_tokens": 2765895.0,
      "step": 1255
    },
    {
      "entropy": 6.348427820205688,
      "epoch": 1.27029752899647,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990624497635152,
      "loss": 6.1468,
      "mean_token_accuracy": 0.119571004062891,
      "num_tokens": 2776897.0,
      "step": 1260
    },
    {
      "entropy": 6.411334657669068,
      "epoch": 1.275340393343419,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499025927896206,
      "loss": 6.1993,
      "mean_token_accuracy": 0.1178061805665493,
      "num_tokens": 2788666.0,
      "step": 1265
    },
    {
      "entropy": 6.403590869903565,
      "epoch": 1.2803832576903682,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989887097511451,
      "loss": 6.0816,
      "mean_token_accuracy": 0.11880902722477912,
      "num_tokens": 2799167.0,
      "step": 1270
    },
    {
      "entropy": 6.2923743724823,
      "epoch": 1.2854261220373173,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989507954440074,
      "loss": 6.0803,
      "mean_token_accuracy": 0.11575406789779663,
      "num_tokens": 2809774.0,
      "step": 1275
    },
    {
      "entropy": 6.41163649559021,
      "epoch": 1.2904689863842662,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989121850926313,
      "loss": 6.1993,
      "mean_token_accuracy": 0.11564520001411438,
      "num_tokens": 2820515.0,
      "step": 1280
    },
    {
      "entropy": 6.407106590270996,
      "epoch": 1.2955118507312153,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988728788170188,
      "loss": 6.1398,
      "mean_token_accuracy": 0.11928385570645332,
      "num_tokens": 2832163.0,
      "step": 1285
    },
    {
      "entropy": 6.310460901260376,
      "epoch": 1.3005547150781644,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988328767393346,
      "loss": 6.1142,
      "mean_token_accuracy": 0.12122111395001411,
      "num_tokens": 2843717.0,
      "step": 1290
    },
    {
      "entropy": 6.301966953277588,
      "epoch": 1.3055975794251133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987921789839061,
      "loss": 6.0834,
      "mean_token_accuracy": 0.1179940603673458,
      "num_tokens": 2854489.0,
      "step": 1295
    },
    {
      "entropy": 6.371503496170044,
      "epoch": 1.3106404437720625,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987507856772229,
      "loss": 6.1238,
      "mean_token_accuracy": 0.12143627852201462,
      "num_tokens": 2865317.0,
      "step": 1300
    },
    {
      "entropy": 6.281271934509277,
      "epoch": 1.3156833081190116,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987086969479364,
      "loss": 6.0504,
      "mean_token_accuracy": 0.1238139659166336,
      "num_tokens": 2877466.0,
      "step": 1305
    },
    {
      "entropy": 6.346838331222534,
      "epoch": 1.3207261724659607,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986659129268592,
      "loss": 6.0828,
      "mean_token_accuracy": 0.1219179905951023,
      "num_tokens": 2887698.0,
      "step": 1310
    },
    {
      "entropy": 6.200007104873658,
      "epoch": 1.3257690368129098,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004986224337469652,
      "loss": 6.0367,
      "mean_token_accuracy": 0.12488842830061912,
      "num_tokens": 2899339.0,
      "step": 1315
    },
    {
      "entropy": 6.343645191192627,
      "epoch": 1.330811901159859,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985782595433886,
      "loss": 6.0781,
      "mean_token_accuracy": 0.12387225553393363,
      "num_tokens": 2908340.0,
      "step": 1320
    },
    {
      "entropy": 6.327138948440552,
      "epoch": 1.3358547655068078,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985333904534237,
      "loss": 6.1782,
      "mean_token_accuracy": 0.12270187363028526,
      "num_tokens": 2920432.0,
      "step": 1325
    },
    {
      "entropy": 6.36779146194458,
      "epoch": 1.340897629853757,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498487826616525,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1162177599966526,
      "num_tokens": 2930385.0,
      "step": 1330
    },
    {
      "entropy": 6.394891881942749,
      "epoch": 1.345940494200706,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004984415681743056,
      "loss": 6.0822,
      "mean_token_accuracy": 0.11857689172029495,
      "num_tokens": 2940760.0,
      "step": 1335
    },
    {
      "entropy": 6.301519727706909,
      "epoch": 1.350983358547655,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983946152705382,
      "loss": 6.0849,
      "mean_token_accuracy": 0.12730747163295747,
      "num_tokens": 2952441.0,
      "step": 1340
    },
    {
      "entropy": 6.3621299266815186,
      "epoch": 1.356026222894604,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983469680511531,
      "loss": 6.1216,
      "mean_token_accuracy": 0.12057074010372162,
      "num_tokens": 2962766.0,
      "step": 1345
    },
    {
      "entropy": 6.23132815361023,
      "epoch": 1.3610690872415532,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004982986266642393,
      "loss": 6.0094,
      "mean_token_accuracy": 0.1286078929901123,
      "num_tokens": 2974239.0,
      "step": 1350
    },
    {
      "entropy": 6.28206877708435,
      "epoch": 1.3661119515885023,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982495912600426,
      "loss": 6.0404,
      "mean_token_accuracy": 0.12513453513383865,
      "num_tokens": 2984497.0,
      "step": 1355
    },
    {
      "entropy": 6.3009498596191404,
      "epoch": 1.3711548159354514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981998619909664,
      "loss": 6.0282,
      "mean_token_accuracy": 0.12919085174798967,
      "num_tokens": 2995367.0,
      "step": 1360
    },
    {
      "entropy": 6.347341632843017,
      "epoch": 1.3761976802824005,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981494390115704,
      "loss": 6.073,
      "mean_token_accuracy": 0.12306225821375846,
      "num_tokens": 3007687.0,
      "step": 1365
    },
    {
      "entropy": 6.3034430027008055,
      "epoch": 1.3812405446293494,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980983224785704,
      "loss": 6.0466,
      "mean_token_accuracy": 0.12317413464188576,
      "num_tokens": 3017505.0,
      "step": 1370
    },
    {
      "entropy": 6.194773483276367,
      "epoch": 1.3862834089762985,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004980465125508376,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1256312608718872,
      "num_tokens": 3028271.0,
      "step": 1375
    },
    {
      "entropy": 6.2702476501464846,
      "epoch": 1.3913262733232477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979940093893986,
      "loss": 6.0498,
      "mean_token_accuracy": 0.1274739347398281,
      "num_tokens": 3040440.0,
      "step": 1380
    },
    {
      "entropy": 6.246666383743286,
      "epoch": 1.3963691376701965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979408131574345,
      "loss": 6.0101,
      "mean_token_accuracy": 0.12436403408646583,
      "num_tokens": 3050999.0,
      "step": 1385
    },
    {
      "entropy": 6.263060140609741,
      "epoch": 1.4014120020171457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978869240202802,
      "loss": 6.0682,
      "mean_token_accuracy": 0.12510961294174194,
      "num_tokens": 3063374.0,
      "step": 1390
    },
    {
      "entropy": 6.368999242782593,
      "epoch": 1.4064548663640948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004978323421454246,
      "loss": 6.094,
      "mean_token_accuracy": 0.12669392451643943,
      "num_tokens": 3075464.0,
      "step": 1395
    },
    {
      "entropy": 6.1546022415161135,
      "epoch": 1.411497730711044,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977770677025092,
      "loss": 5.9634,
      "mean_token_accuracy": 0.12526423931121827,
      "num_tokens": 3086398.0,
      "step": 1400
    },
    {
      "entropy": 6.278340196609497,
      "epoch": 1.416540595057993,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004977211008633284,
      "loss": 6.1006,
      "mean_token_accuracy": 0.11808581873774529,
      "num_tokens": 3097777.0,
      "step": 1405
    },
    {
      "entropy": 6.323656940460205,
      "epoch": 1.4215834594049421,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976644418018285,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1321638911962509,
      "num_tokens": 3108711.0,
      "step": 1410
    },
    {
      "entropy": 6.269476699829101,
      "epoch": 1.426626323751891,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497607090694107,
      "loss": 6.0955,
      "mean_token_accuracy": 0.1233494944870472,
      "num_tokens": 3119589.0,
      "step": 1415
    },
    {
      "entropy": 6.261565446853638,
      "epoch": 1.4316691880988401,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004975490477184126,
      "loss": 6.0183,
      "mean_token_accuracy": 0.13225764781236649,
      "num_tokens": 3131744.0,
      "step": 1420
    },
    {
      "entropy": 6.313161945343017,
      "epoch": 1.4367120524457893,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497490313055144,
      "loss": 6.0621,
      "mean_token_accuracy": 0.12190337777137757,
      "num_tokens": 3143526.0,
      "step": 1425
    },
    {
      "entropy": 6.223160982131958,
      "epoch": 1.4417549167927382,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004974308868868501,
      "loss": 5.9978,
      "mean_token_accuracy": 0.12351568043231964,
      "num_tokens": 3154561.0,
      "step": 1430
    },
    {
      "entropy": 6.3151466846466064,
      "epoch": 1.4467977811396873,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973707693982287,
      "loss": 6.1643,
      "mean_token_accuracy": 0.11945203095674514,
      "num_tokens": 3166052.0,
      "step": 1435
    },
    {
      "entropy": 6.319597864151001,
      "epoch": 1.4518406454866364,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004973099607761264,
      "loss": 6.1199,
      "mean_token_accuracy": 0.12031879499554635,
      "num_tokens": 3179335.0,
      "step": 1440
    },
    {
      "entropy": 6.243300199508667,
      "epoch": 1.4568835098335855,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972484612095378,
      "loss": 6.0054,
      "mean_token_accuracy": 0.12574711814522743,
      "num_tokens": 3192129.0,
      "step": 1445
    },
    {
      "entropy": 6.348361444473267,
      "epoch": 1.4619263741805346,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497186270889605,
      "loss": 6.1172,
      "mean_token_accuracy": 0.1257016494870186,
      "num_tokens": 3202792.0,
      "step": 1450
    },
    {
      "entropy": 6.289998245239258,
      "epoch": 1.4669692385274837,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004971233900096169,
      "loss": 6.0474,
      "mean_token_accuracy": 0.12485183402895927,
      "num_tokens": 3213707.0,
      "step": 1455
    },
    {
      "entropy": 6.2773839950561525,
      "epoch": 1.4720121028744326,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004970598187650085,
      "loss": 6.0406,
      "mean_token_accuracy": 0.12844547405838966,
      "num_tokens": 3224531.0,
      "step": 1460
    },
    {
      "entropy": 6.3069030284881595,
      "epoch": 1.4770549672213817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496995557353361,
      "loss": 6.0255,
      "mean_token_accuracy": 0.13030240088701248,
      "num_tokens": 3234656.0,
      "step": 1465
    },
    {
      "entropy": 6.21710901260376,
      "epoch": 1.4820978315683309,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969306059744004,
      "loss": 6.0545,
      "mean_token_accuracy": 0.12597464323043822,
      "num_tokens": 3246745.0,
      "step": 1470
    },
    {
      "entropy": 6.216967058181763,
      "epoch": 1.4871406959152798,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968649648299968,
      "loss": 6.0314,
      "mean_token_accuracy": 0.12574660703539847,
      "num_tokens": 3257277.0,
      "step": 1475
    },
    {
      "entropy": 6.296077060699463,
      "epoch": 1.4921835602622289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967986341241646,
      "loss": 6.001,
      "mean_token_accuracy": 0.1272491253912449,
      "num_tokens": 3267704.0,
      "step": 1480
    },
    {
      "entropy": 6.2856720924377445,
      "epoch": 1.497226424609178,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967316140630611,
      "loss": 6.0725,
      "mean_token_accuracy": 0.12544610872864723,
      "num_tokens": 3279772.0,
      "step": 1485
    },
    {
      "entropy": 6.18753604888916,
      "epoch": 1.502269288956127,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966639048549862,
      "loss": 5.9753,
      "mean_token_accuracy": 0.1282753072679043,
      "num_tokens": 3290536.0,
      "step": 1490
    },
    {
      "entropy": 6.208110904693603,
      "epoch": 1.5073121533030762,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004965955067103816,
      "loss": 6.0397,
      "mean_token_accuracy": 0.13054657354950905,
      "num_tokens": 3302051.0,
      "step": 1495
    },
    {
      "entropy": 6.247125911712646,
      "epoch": 1.5123550176500253,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965264198418305,
      "loss": 6.0154,
      "mean_token_accuracy": 0.11876789480447769,
      "num_tokens": 3314324.0,
      "step": 1500
    },
    {
      "epoch": 1.5123550176500253,
      "eval_entropy": 6.089648048386319,
      "eval_loss": 6.147760391235352,
      "eval_mean_token_accuracy": 0.12741849596112131,
      "eval_num_tokens": 3314324.0,
      "eval_runtime": 2.0304,
      "eval_samples_per_second": 1545.537,
      "eval_steps_per_second": 193.561,
      "step": 1500
    },
    {
      "entropy": 6.156036901473999,
      "epoch": 1.5173978819969742,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004964566444640564,
      "loss": 6.0108,
      "mean_token_accuracy": 0.12939631938934326,
      "num_tokens": 3326246.0,
      "step": 1505
    },
    {
      "entropy": 6.238477659225464,
      "epoch": 1.5224407463439233,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963861807939227,
      "loss": 6.0146,
      "mean_token_accuracy": 0.12993920966982841,
      "num_tokens": 3336308.0,
      "step": 1510
    },
    {
      "entropy": 6.137201929092408,
      "epoch": 1.5274836106908725,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963150290504321,
      "loss": 5.9478,
      "mean_token_accuracy": 0.13382459282875062,
      "num_tokens": 3347596.0,
      "step": 1515
    },
    {
      "entropy": 6.140363264083862,
      "epoch": 1.5325264750378214,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004962431894547259,
      "loss": 5.9261,
      "mean_token_accuracy": 0.1316240534186363,
      "num_tokens": 3358370.0,
      "step": 1520
    },
    {
      "entropy": 6.282736682891846,
      "epoch": 1.5375693393847705,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961706622300834,
      "loss": 5.9539,
      "mean_token_accuracy": 0.12983666807413102,
      "num_tokens": 3368630.0,
      "step": 1525
    },
    {
      "entropy": 6.224072694778442,
      "epoch": 1.5426122037317196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960974476019206,
      "loss": 5.964,
      "mean_token_accuracy": 0.13069895058870315,
      "num_tokens": 3379952.0,
      "step": 1530
    },
    {
      "entropy": 6.137469005584717,
      "epoch": 1.5476550680786687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960235457977905,
      "loss": 5.9064,
      "mean_token_accuracy": 0.1311415933072567,
      "num_tokens": 3390097.0,
      "step": 1535
    },
    {
      "entropy": 6.11315426826477,
      "epoch": 1.5526979324256178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959489570473814,
      "loss": 5.9007,
      "mean_token_accuracy": 0.13429841324687003,
      "num_tokens": 3399778.0,
      "step": 1540
    },
    {
      "entropy": 6.219993686676025,
      "epoch": 1.557740796772567,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000495873681582517,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1224518284201622,
      "num_tokens": 3410213.0,
      "step": 1545
    },
    {
      "entropy": 6.288167142868042,
      "epoch": 1.5627836611195158,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957977196371552,
      "loss": 6.0564,
      "mean_token_accuracy": 0.12722626626491546,
      "num_tokens": 3422071.0,
      "step": 1550
    },
    {
      "entropy": 6.228869485855102,
      "epoch": 1.567826525466465,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004957210714473873,
      "loss": 6.024,
      "mean_token_accuracy": 0.12989916950464248,
      "num_tokens": 3432102.0,
      "step": 1555
    },
    {
      "entropy": 6.2041120529174805,
      "epoch": 1.572869389813414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004956437372514378,
      "loss": 6.0039,
      "mean_token_accuracy": 0.12639209032058715,
      "num_tokens": 3442821.0,
      "step": 1560
    },
    {
      "entropy": 6.140823221206665,
      "epoch": 1.577912254160363,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004955657172896631,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1277303285896778,
      "num_tokens": 3454378.0,
      "step": 1565
    },
    {
      "entropy": 6.172880506515503,
      "epoch": 1.582955118507312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004954870118045511,
      "loss": 5.9748,
      "mean_token_accuracy": 0.13318561986088753,
      "num_tokens": 3463933.0,
      "step": 1570
    },
    {
      "entropy": 6.177058744430542,
      "epoch": 1.5879979828542612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954076210407203,
      "loss": 5.9138,
      "mean_token_accuracy": 0.12689697593450547,
      "num_tokens": 3473993.0,
      "step": 1575
    },
    {
      "entropy": 6.158699464797974,
      "epoch": 1.5930408472012103,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953275452449188,
      "loss": 6.0649,
      "mean_token_accuracy": 0.12153254449367523,
      "num_tokens": 3484882.0,
      "step": 1580
    },
    {
      "entropy": 6.223286962509155,
      "epoch": 1.5980837115481594,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004952467846660243,
      "loss": 5.9822,
      "mean_token_accuracy": 0.12474498450756073,
      "num_tokens": 3496503.0,
      "step": 1585
    },
    {
      "entropy": 6.219234228134155,
      "epoch": 1.6031265758951085,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004951653395550423,
      "loss": 6.0554,
      "mean_token_accuracy": 0.12866610437631607,
      "num_tokens": 3508696.0,
      "step": 1590
    },
    {
      "entropy": 6.2126219272613525,
      "epoch": 1.6081694402420574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950832101651062,
      "loss": 5.9758,
      "mean_token_accuracy": 0.13273679241538047,
      "num_tokens": 3519542.0,
      "step": 1595
    },
    {
      "entropy": 6.218908977508545,
      "epoch": 1.6132123045890066,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950003967514762,
      "loss": 5.9417,
      "mean_token_accuracy": 0.12883660346269607,
      "num_tokens": 3531576.0,
      "step": 1600
    },
    {
      "entropy": 6.181266021728516,
      "epoch": 1.6182551689359557,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000494916899571538,
      "loss": 5.905,
      "mean_token_accuracy": 0.13329532369971275,
      "num_tokens": 3542816.0,
      "step": 1605
    },
    {
      "entropy": 6.208526515960694,
      "epoch": 1.6232980332829046,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948327188848029,
      "loss": 6.0416,
      "mean_token_accuracy": 0.1266243115067482,
      "num_tokens": 3554100.0,
      "step": 1610
    },
    {
      "entropy": 6.285739994049072,
      "epoch": 1.6283408976298537,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004947478549529065,
      "loss": 5.8826,
      "mean_token_accuracy": 0.12605574503540992,
      "num_tokens": 3565048.0,
      "step": 1615
    },
    {
      "entropy": 6.161268329620361,
      "epoch": 1.6333837619768028,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946623080396077,
      "loss": 5.9799,
      "mean_token_accuracy": 0.12972840890288354,
      "num_tokens": 3575587.0,
      "step": 1620
    },
    {
      "entropy": 6.198655223846435,
      "epoch": 1.638426626323752,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004945760784107884,
      "loss": 6.0417,
      "mean_token_accuracy": 0.1309143140912056,
      "num_tokens": 3587059.0,
      "step": 1625
    },
    {
      "entropy": 6.121437358856201,
      "epoch": 1.643469490670701,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004944891663344522,
      "loss": 5.8684,
      "mean_token_accuracy": 0.13101207837462425,
      "num_tokens": 3598492.0,
      "step": 1630
    },
    {
      "entropy": 6.1245574951171875,
      "epoch": 1.6485123550176501,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004944015720807239,
      "loss": 5.9395,
      "mean_token_accuracy": 0.12826012969017028,
      "num_tokens": 3609513.0,
      "step": 1635
    },
    {
      "entropy": 6.186050367355347,
      "epoch": 1.653555219364599,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004943132959218484,
      "loss": 5.9739,
      "mean_token_accuracy": 0.13249149844050406,
      "num_tokens": 3621327.0,
      "step": 1640
    },
    {
      "entropy": 6.106121635437011,
      "epoch": 1.6585980837115482,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049422433813219,
      "loss": 5.887,
      "mean_token_accuracy": 0.13323233276605606,
      "num_tokens": 3632136.0,
      "step": 1645
    },
    {
      "entropy": 6.168408966064453,
      "epoch": 1.6636409480584973,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004941346989882318,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1332289457321167,
      "num_tokens": 3642271.0,
      "step": 1650
    },
    {
      "entropy": 6.016178035736084,
      "epoch": 1.6686838124054462,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004940443787685739,
      "loss": 5.9801,
      "mean_token_accuracy": 0.12791188880801202,
      "num_tokens": 3652355.0,
      "step": 1655
    },
    {
      "entropy": 6.278066539764405,
      "epoch": 1.6737266767523953,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004939533777539341,
      "loss": 5.9537,
      "mean_token_accuracy": 0.13114508166909217,
      "num_tokens": 3663714.0,
      "step": 1660
    },
    {
      "entropy": 6.035008478164673,
      "epoch": 1.6787695410993444,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004938616962271453,
      "loss": 5.9067,
      "mean_token_accuracy": 0.1274005278944969,
      "num_tokens": 3674675.0,
      "step": 1665
    },
    {
      "entropy": 6.157523965835571,
      "epoch": 1.6838124054462935,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004937693344731561,
      "loss": 5.9846,
      "mean_token_accuracy": 0.134155460447073,
      "num_tokens": 3686403.0,
      "step": 1670
    },
    {
      "entropy": 6.123535585403443,
      "epoch": 1.6888552697932426,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004936762927790286,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14105360731482505,
      "num_tokens": 3697259.0,
      "step": 1675
    },
    {
      "entropy": 6.131226682662964,
      "epoch": 1.6938981341401917,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935825714339389,
      "loss": 5.9318,
      "mean_token_accuracy": 0.13263342157006264,
      "num_tokens": 3708536.0,
      "step": 1680
    },
    {
      "entropy": 6.21954231262207,
      "epoch": 1.6989409984871406,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004934881707291748,
      "loss": 5.9456,
      "mean_token_accuracy": 0.13225506469607354,
      "num_tokens": 3719181.0,
      "step": 1685
    },
    {
      "entropy": 6.048918867111206,
      "epoch": 1.7039838628340898,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004933930909581362,
      "loss": 5.7999,
      "mean_token_accuracy": 0.13783433213829993,
      "num_tokens": 3729325.0,
      "step": 1690
    },
    {
      "entropy": 6.120263910293579,
      "epoch": 1.7090267271810389,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004932973324163328,
      "loss": 5.9244,
      "mean_token_accuracy": 0.12808132991194726,
      "num_tokens": 3740462.0,
      "step": 1695
    },
    {
      "entropy": 6.212024211883545,
      "epoch": 1.7140695915279878,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004932008954013846,
      "loss": 6.0037,
      "mean_token_accuracy": 0.13034727275371552,
      "num_tokens": 3752523.0,
      "step": 1700
    },
    {
      "entropy": 6.034075117111206,
      "epoch": 1.7191124558749369,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931037802130201,
      "loss": 5.8662,
      "mean_token_accuracy": 0.13170775398612022,
      "num_tokens": 3764120.0,
      "step": 1705
    },
    {
      "entropy": 6.1746021747589115,
      "epoch": 1.724155320221886,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930059871530755,
      "loss": 5.9938,
      "mean_token_accuracy": 0.12606719359755517,
      "num_tokens": 3776733.0,
      "step": 1710
    },
    {
      "entropy": 6.14895544052124,
      "epoch": 1.7291981845688351,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929075165254938,
      "loss": 5.969,
      "mean_token_accuracy": 0.13328379690647124,
      "num_tokens": 3787799.0,
      "step": 1715
    },
    {
      "entropy": 6.066791725158692,
      "epoch": 1.7342410489157842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004928083686363241,
      "loss": 5.8874,
      "mean_token_accuracy": 0.13170522972941398,
      "num_tokens": 3797995.0,
      "step": 1720
    },
    {
      "entropy": 6.203559875488281,
      "epoch": 1.7392839132627333,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004927085437937204,
      "loss": 5.9704,
      "mean_token_accuracy": 0.1316309779882431,
      "num_tokens": 3809580.0,
      "step": 1725
    },
    {
      "entropy": 6.136819458007812,
      "epoch": 1.7443267776096822,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926080423079404,
      "loss": 6.002,
      "mean_token_accuracy": 0.13009854927659034,
      "num_tokens": 3820150.0,
      "step": 1730
    },
    {
      "entropy": 6.137077331542969,
      "epoch": 1.7493696419566314,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492506864491345,
      "loss": 5.8522,
      "mean_token_accuracy": 0.1314531646668911,
      "num_tokens": 3830391.0,
      "step": 1735
    },
    {
      "entropy": 6.01658182144165,
      "epoch": 1.7544125063035805,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004924050106583976,
      "loss": 5.8555,
      "mean_token_accuracy": 0.13897303864359856,
      "num_tokens": 3842773.0,
      "step": 1740
    },
    {
      "entropy": 6.169336891174316,
      "epoch": 1.7594553706505294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923024811256618,
      "loss": 5.9072,
      "mean_token_accuracy": 0.13076598420739174,
      "num_tokens": 3853646.0,
      "step": 1745
    },
    {
      "entropy": 6.141702365875244,
      "epoch": 1.7644982349974785,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492199276211802,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1321739748120308,
      "num_tokens": 3865319.0,
      "step": 1750
    },
    {
      "entropy": 6.057036447525024,
      "epoch": 1.7695410993444276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004920953962375813,
      "loss": 5.9197,
      "mean_token_accuracy": 0.12733314409852028,
      "num_tokens": 3877107.0,
      "step": 1755
    },
    {
      "entropy": 6.127471590042115,
      "epoch": 1.7745839636913767,
      "grad_norm": 1.0,
      "learning_rate": 0.0004919908415258612,
      "loss": 5.9594,
      "mean_token_accuracy": 0.1326076626777649,
      "num_tokens": 3889334.0,
      "step": 1760
    },
    {
      "entropy": 6.109628105163575,
      "epoch": 1.7796268280383258,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004918856124016002,
      "loss": 5.8521,
      "mean_token_accuracy": 0.13769446834921836,
      "num_tokens": 3899482.0,
      "step": 1765
    },
    {
      "entropy": 6.055948114395141,
      "epoch": 1.784669692385275,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917797091918526,
      "loss": 5.7791,
      "mean_token_accuracy": 0.13959548696875573,
      "num_tokens": 3910216.0,
      "step": 1770
    },
    {
      "entropy": 6.133174705505371,
      "epoch": 1.789712556732224,
      "grad_norm": 0.875,
      "learning_rate": 0.0004916731322257683,
      "loss": 5.9456,
      "mean_token_accuracy": 0.1310464061796665,
      "num_tokens": 3922246.0,
      "step": 1775
    },
    {
      "entropy": 6.121117448806762,
      "epoch": 1.794755421079173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004915658818345908,
      "loss": 5.8371,
      "mean_token_accuracy": 0.14053154289722442,
      "num_tokens": 3933688.0,
      "step": 1780
    },
    {
      "entropy": 6.121936845779419,
      "epoch": 1.799798285426122,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004914579583516568,
      "loss": 5.927,
      "mean_token_accuracy": 0.13004440814256668,
      "num_tokens": 3944168.0,
      "step": 1785
    },
    {
      "entropy": 6.069381046295166,
      "epoch": 1.804841149773071,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491349362112395,
      "loss": 5.8582,
      "mean_token_accuracy": 0.13762850761413575,
      "num_tokens": 3954346.0,
      "step": 1790
    },
    {
      "entropy": 6.156457567214966,
      "epoch": 1.80988401412002,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000491240093454325,
      "loss": 5.9712,
      "mean_token_accuracy": 0.12970432341098787,
      "num_tokens": 3966375.0,
      "step": 1795
    },
    {
      "entropy": 6.138356399536133,
      "epoch": 1.8149268784669692,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911301527170563,
      "loss": 5.9372,
      "mean_token_accuracy": 0.13539533764123918,
      "num_tokens": 3977030.0,
      "step": 1800
    },
    {
      "entropy": 6.0566565990448,
      "epoch": 1.8199697428139183,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004910195402422873,
      "loss": 5.8489,
      "mean_token_accuracy": 0.13358401656150817,
      "num_tokens": 3988064.0,
      "step": 1805
    },
    {
      "entropy": 6.10338978767395,
      "epoch": 1.8250126071608674,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004909082563738042,
      "loss": 5.8977,
      "mean_token_accuracy": 0.13457708284258843,
      "num_tokens": 3997994.0,
      "step": 1810
    },
    {
      "entropy": 6.171204137802124,
      "epoch": 1.8300554715078166,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004907963014574797,
      "loss": 5.9426,
      "mean_token_accuracy": 0.12793719545006751,
      "num_tokens": 4008645.0,
      "step": 1815
    },
    {
      "entropy": 6.067741918563843,
      "epoch": 1.8350983358547657,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004906836758412722,
      "loss": 5.891,
      "mean_token_accuracy": 0.13499666452407838,
      "num_tokens": 4020387.0,
      "step": 1820
    },
    {
      "entropy": 5.933810758590698,
      "epoch": 1.8401412002017146,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905703798752248,
      "loss": 5.7667,
      "mean_token_accuracy": 0.1390545152127743,
      "num_tokens": 4032720.0,
      "step": 1825
    },
    {
      "entropy": 6.113948631286621,
      "epoch": 1.8451840645486637,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000490456413911464,
      "loss": 5.849,
      "mean_token_accuracy": 0.13772407844662665,
      "num_tokens": 4043438.0,
      "step": 1830
    },
    {
      "entropy": 5.984061193466187,
      "epoch": 1.8502269288956126,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004903417783041987,
      "loss": 5.8149,
      "mean_token_accuracy": 0.14559042751789092,
      "num_tokens": 4054467.0,
      "step": 1835
    },
    {
      "entropy": 6.068659162521362,
      "epoch": 1.8552697932425617,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004902264734097188,
      "loss": 5.8142,
      "mean_token_accuracy": 0.13721002638339996,
      "num_tokens": 4066418.0,
      "step": 1840
    },
    {
      "entropy": 6.145805835723877,
      "epoch": 1.8603126575895108,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004901104995863948,
      "loss": 5.9205,
      "mean_token_accuracy": 0.13175863698124884,
      "num_tokens": 4078770.0,
      "step": 1845
    },
    {
      "entropy": 6.075065279006958,
      "epoch": 1.86535552193646,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004899938571946759,
      "loss": 5.9581,
      "mean_token_accuracy": 0.12940308153629304,
      "num_tokens": 4089998.0,
      "step": 1850
    },
    {
      "entropy": 6.088465833663941,
      "epoch": 1.870398386283409,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004898765465970894,
      "loss": 5.8615,
      "mean_token_accuracy": 0.13679955005645753,
      "num_tokens": 4101409.0,
      "step": 1855
    },
    {
      "entropy": 6.117992305755616,
      "epoch": 1.8754412506303582,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004897585681582393,
      "loss": 5.9812,
      "mean_token_accuracy": 0.12942875251173974,
      "num_tokens": 4113598.0,
      "step": 1860
    },
    {
      "entropy": 6.047978687286377,
      "epoch": 1.8804841149773073,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004896399222448053,
      "loss": 5.9059,
      "mean_token_accuracy": 0.13164842873811722,
      "num_tokens": 4124639.0,
      "step": 1865
    },
    {
      "entropy": 6.22193021774292,
      "epoch": 1.8855269793242562,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004895206092255417,
      "loss": 5.9702,
      "mean_token_accuracy": 0.12647129818797112,
      "num_tokens": 4136280.0,
      "step": 1870
    },
    {
      "entropy": 6.056543493270874,
      "epoch": 1.8905698436712053,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004894006294712761,
      "loss": 5.8227,
      "mean_token_accuracy": 0.1319723516702652,
      "num_tokens": 4148251.0,
      "step": 1875
    },
    {
      "entropy": 6.069836711883545,
      "epoch": 1.8956127080181542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004892799833549083,
      "loss": 5.9088,
      "mean_token_accuracy": 0.13684113174676896,
      "num_tokens": 4159298.0,
      "step": 1880
    },
    {
      "entropy": 6.108313369750976,
      "epoch": 1.9006555723651033,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004891586712514091,
      "loss": 5.8093,
      "mean_token_accuracy": 0.13893356025218964,
      "num_tokens": 4169974.0,
      "step": 1885
    },
    {
      "entropy": 6.052180051803589,
      "epoch": 1.9056984367120524,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004890366935378196,
      "loss": 5.8247,
      "mean_token_accuracy": 0.14235706329345704,
      "num_tokens": 4181442.0,
      "step": 1890
    },
    {
      "entropy": 6.107061386108398,
      "epoch": 1.9107413010590015,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004889140505932492,
      "loss": 5.8668,
      "mean_token_accuracy": 0.13770687878131865,
      "num_tokens": 4191535.0,
      "step": 1895
    },
    {
      "entropy": 6.075644683837891,
      "epoch": 1.9157841654059506,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004887907427988749,
      "loss": 5.8944,
      "mean_token_accuracy": 0.13668857961893083,
      "num_tokens": 4201741.0,
      "step": 1900
    },
    {
      "entropy": 6.0178862571716305,
      "epoch": 1.9208270297528998,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004886667705379403,
      "loss": 5.7844,
      "mean_token_accuracy": 0.13908884152770043,
      "num_tokens": 4211762.0,
      "step": 1905
    },
    {
      "entropy": 6.073857307434082,
      "epoch": 1.9258698940998489,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004885421341957541,
      "loss": 5.8684,
      "mean_token_accuracy": 0.1342634789645672,
      "num_tokens": 4222322.0,
      "step": 1910
    },
    {
      "entropy": 6.1019774913787845,
      "epoch": 1.9309127584467978,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004884168341596888,
      "loss": 5.9438,
      "mean_token_accuracy": 0.13611835464835167,
      "num_tokens": 4232166.0,
      "step": 1915
    },
    {
      "entropy": 6.0883242130279545,
      "epoch": 1.9359556227937469,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004882908708191797,
      "loss": 5.8871,
      "mean_token_accuracy": 0.1351541370153427,
      "num_tokens": 4243651.0,
      "step": 1920
    },
    {
      "entropy": 6.0817262649536135,
      "epoch": 1.9409984871406958,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048816424456572403,
      "loss": 5.814,
      "mean_token_accuracy": 0.14198361784219743,
      "num_tokens": 4255137.0,
      "step": 1925
    },
    {
      "entropy": 5.938052940368652,
      "epoch": 1.946041351487645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048803695579287905,
      "loss": 5.8279,
      "mean_token_accuracy": 0.13684711009263992,
      "num_tokens": 4264904.0,
      "step": 1930
    },
    {
      "entropy": 6.035186624526977,
      "epoch": 1.951084215834594,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048790900489626104,
      "loss": 5.7888,
      "mean_token_accuracy": 0.13858872056007385,
      "num_tokens": 4275305.0,
      "step": 1935
    },
    {
      "entropy": 6.053544712066651,
      "epoch": 1.9561270801815431,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004877803922735445,
      "loss": 5.8805,
      "mean_token_accuracy": 0.13514195308089255,
      "num_tokens": 4287038.0,
      "step": 1940
    },
    {
      "entropy": 6.167809200286865,
      "epoch": 1.9611699445284922,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004876511183244603,
      "loss": 5.9877,
      "mean_token_accuracy": 0.12615374028682708,
      "num_tokens": 4298668.0,
      "step": 1945
    },
    {
      "entropy": 5.978178691864014,
      "epoch": 1.9662128088754414,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048752118345079496,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1383742056787014,
      "num_tokens": 4309429.0,
      "step": 1950
    },
    {
      "entropy": 6.073020601272583,
      "epoch": 1.9712556732223905,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048739058805638905,
      "loss": 5.8624,
      "mean_token_accuracy": 0.1353800967335701,
      "num_tokens": 4321137.0,
      "step": 1955
    },
    {
      "entropy": 6.074612426757812,
      "epoch": 1.9762985375693394,
      "grad_norm": 0.9375,
      "learning_rate": 0.000487259332547136,
      "loss": 5.8661,
      "mean_token_accuracy": 0.12799250036478044,
      "num_tokens": 4331249.0,
      "step": 1960
    },
    {
      "entropy": 6.084528923034668,
      "epoch": 1.9813414019162885,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000487127417330981,
      "loss": 5.8556,
      "mean_token_accuracy": 0.13444280624389648,
      "num_tokens": 4342154.0,
      "step": 1965
    },
    {
      "entropy": 6.1047789573669435,
      "epoch": 1.9863842662632374,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004869948428179196,
      "loss": 5.9024,
      "mean_token_accuracy": 0.13036157563328743,
      "num_tokens": 4353906.0,
      "step": 1970
    },
    {
      "entropy": 6.083949089050293,
      "epoch": 1.9914271306101865,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004868616094199964,
      "loss": 5.9485,
      "mean_token_accuracy": 0.1291046179831028,
      "num_tokens": 4366416.0,
      "step": 1975
    },
    {
      "entropy": 6.076101493835449,
      "epoch": 1.9964699949571356,
      "grad_norm": 1.03125,
      "learning_rate": 0.000486727717551304,
      "loss": 5.8428,
      "mean_token_accuracy": 0.13770969063043595,
      "num_tokens": 4376743.0,
      "step": 1980
    },
    {
      "entropy": 5.963961230383979,
      "epoch": 2.00100857286939,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004865931676279813,
      "loss": 5.636,
      "mean_token_accuracy": 0.14866483873791164,
      "num_tokens": 4385379.0,
      "step": 1985
    },
    {
      "entropy": 6.011814737319947,
      "epoch": 2.006051437216339,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004864579600682125,
      "loss": 5.5817,
      "mean_token_accuracy": 0.14293505102396012,
      "num_tokens": 4396872.0,
      "step": 1990
    },
    {
      "entropy": 6.032746171951294,
      "epoch": 2.011094301563288,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048632209529222607,
      "loss": 5.5241,
      "mean_token_accuracy": 0.14778304547071458,
      "num_tokens": 4407075.0,
      "step": 1995
    },
    {
      "entropy": 5.942056798934937,
      "epoch": 2.016137165910237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048618557372229253,
      "loss": 5.5591,
      "mean_token_accuracy": 0.15164274126291274,
      "num_tokens": 4418020.0,
      "step": 2000
    },
    {
      "epoch": 2.016137165910237,
      "eval_entropy": 5.723532737363082,
      "eval_loss": 5.9471025466918945,
      "eval_mean_token_accuracy": 0.1362081553796472,
      "eval_num_tokens": 4418020.0,
      "eval_runtime": 2.045,
      "eval_samples_per_second": 1534.452,
      "eval_steps_per_second": 192.173,
      "step": 2000
    },
    {
      "entropy": 5.940024185180664,
      "epoch": 2.0211800302571863,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004860483957827243,
      "loss": 5.5724,
      "mean_token_accuracy": 0.14962393194437026,
      "num_tokens": 4429740.0,
      "step": 2005
    },
    {
      "entropy": 6.0355141162872314,
      "epoch": 2.026222894604135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004859105618998735,
      "loss": 5.5662,
      "mean_token_accuracy": 0.14324932247400285,
      "num_tokens": 4440183.0,
      "step": 2010
    },
    {
      "entropy": 5.9932300567626955,
      "epoch": 2.031265758951084,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004857720725021311,
      "loss": 5.5371,
      "mean_token_accuracy": 0.14567766189575196,
      "num_tokens": 4451750.0,
      "step": 2015
    },
    {
      "entropy": 5.812967920303345,
      "epoch": 2.036308623298033,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048563292801992534,
      "loss": 5.4972,
      "mean_token_accuracy": 0.14970174431800842,
      "num_tokens": 4462071.0,
      "step": 2020
    },
    {
      "entropy": 6.048516750335693,
      "epoch": 2.0413514876449823,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004854931288857203,
      "loss": 5.6861,
      "mean_token_accuracy": 0.1418111339211464,
      "num_tokens": 4474241.0,
      "step": 2025
    },
    {
      "entropy": 5.946160411834716,
      "epoch": 2.0463943519919314,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048535267553401505,
      "loss": 5.5198,
      "mean_token_accuracy": 0.14917661547660827,
      "num_tokens": 4487083.0,
      "step": 2030
    },
    {
      "entropy": 5.8958487033844,
      "epoch": 2.0514372163388805,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048521156840134193,
      "loss": 5.5706,
      "mean_token_accuracy": 0.1516394168138504,
      "num_tokens": 4497181.0,
      "step": 2035
    },
    {
      "entropy": 5.988409042358398,
      "epoch": 2.0564800806858297,
      "grad_norm": 1.0625,
      "learning_rate": 0.000485069807926265,
      "loss": 5.5509,
      "mean_token_accuracy": 0.14955907613039016,
      "num_tokens": 4508250.0,
      "step": 2040
    },
    {
      "entropy": 5.9201305389404295,
      "epoch": 2.0615229450327788,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004849273945493792,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15315290838479995,
      "num_tokens": 4519778.0,
      "step": 2045
    },
    {
      "entropy": 5.911941957473755,
      "epoch": 2.066565809379728,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004847843287133084,
      "loss": 5.5338,
      "mean_token_accuracy": 0.14484044089913367,
      "num_tokens": 4530879.0,
      "step": 2050
    },
    {
      "entropy": 5.95937180519104,
      "epoch": 2.0716086737266766,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004846406108627048,
      "loss": 5.6061,
      "mean_token_accuracy": 0.15488057881593703,
      "num_tokens": 4542560.0,
      "step": 2055
    },
    {
      "entropy": 5.874370336532593,
      "epoch": 2.0766515380736257,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048449624144424665,
      "loss": 5.504,
      "mean_token_accuracy": 0.15539904236793517,
      "num_tokens": 4552213.0,
      "step": 2060
    },
    {
      "entropy": 5.914573717117309,
      "epoch": 2.081694402420575,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048435122090663744,
      "loss": 5.5253,
      "mean_token_accuracy": 0.15514215230941772,
      "num_tokens": 4562798.0,
      "step": 2065
    },
    {
      "entropy": 5.906200838088989,
      "epoch": 2.086737266767524,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048420554970060424,
      "loss": 5.5632,
      "mean_token_accuracy": 0.142371766269207,
      "num_tokens": 4574534.0,
      "step": 2070
    },
    {
      "entropy": 5.931839179992676,
      "epoch": 2.091780131114473,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048405922827889677,
      "loss": 5.5342,
      "mean_token_accuracy": 0.14170490056276322,
      "num_tokens": 4586239.0,
      "step": 2075
    },
    {
      "entropy": 5.883896541595459,
      "epoch": 2.096822995461422,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048391225709628515,
      "loss": 5.5777,
      "mean_token_accuracy": 0.14845428690314294,
      "num_tokens": 4598018.0,
      "step": 2080
    },
    {
      "entropy": 5.90255184173584,
      "epoch": 2.1018658598083713,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004837646366095593,
      "loss": 5.5498,
      "mean_token_accuracy": 0.14273814857006073,
      "num_tokens": 4608985.0,
      "step": 2085
    },
    {
      "entropy": 5.972240877151489,
      "epoch": 2.1069087241553204,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048361636727752716,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1534451276063919,
      "num_tokens": 4621103.0,
      "step": 2090
    },
    {
      "entropy": 5.869103240966797,
      "epoch": 2.1119515885022695,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004834674495610131,
      "loss": 5.5826,
      "mean_token_accuracy": 0.1498872831463814,
      "num_tokens": 4632244.0,
      "step": 2095
    },
    {
      "entropy": 5.895373868942261,
      "epoch": 2.116994452849218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048331788392285693,
      "loss": 5.5021,
      "mean_token_accuracy": 0.15359969437122345,
      "num_tokens": 4642387.0,
      "step": 2100
    },
    {
      "entropy": 5.9281104564666744,
      "epoch": 2.1220373171961673,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048316767082791205,
      "loss": 5.6388,
      "mean_token_accuracy": 0.14638803526759148,
      "num_tokens": 4655181.0,
      "step": 2105
    },
    {
      "entropy": 5.84404764175415,
      "epoch": 2.1270801815431164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048301681074304427,
      "loss": 5.5393,
      "mean_token_accuracy": 0.14985166043043135,
      "num_tokens": 4665664.0,
      "step": 2110
    },
    {
      "entropy": 5.846418714523315,
      "epoch": 2.1321230458900655,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048286530413713016,
      "loss": 5.5091,
      "mean_token_accuracy": 0.15592919886112214,
      "num_tokens": 4677683.0,
      "step": 2115
    },
    {
      "entropy": 5.902777290344238,
      "epoch": 2.1371659102370146,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048271315148105574,
      "loss": 5.5582,
      "mean_token_accuracy": 0.1461687430739403,
      "num_tokens": 4688891.0,
      "step": 2120
    },
    {
      "entropy": 5.846302080154419,
      "epoch": 2.1422087745839637,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000482560353247715,
      "loss": 5.5815,
      "mean_token_accuracy": 0.14932963848114014,
      "num_tokens": 4700765.0,
      "step": 2125
    },
    {
      "entropy": 5.894233274459839,
      "epoch": 2.147251638930913,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004824069099120084,
      "loss": 5.5988,
      "mean_token_accuracy": 0.14528623223304749,
      "num_tokens": 4712004.0,
      "step": 2130
    },
    {
      "entropy": 5.949212646484375,
      "epoch": 2.152294503277862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004822528219508413,
      "loss": 5.5182,
      "mean_token_accuracy": 0.15119172781705856,
      "num_tokens": 4722427.0,
      "step": 2135
    },
    {
      "entropy": 5.893136692047119,
      "epoch": 2.157337367624811,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004820980898431227,
      "loss": 5.5911,
      "mean_token_accuracy": 0.15657991245388986,
      "num_tokens": 4733815.0,
      "step": 2140
    },
    {
      "entropy": 5.882743215560913,
      "epoch": 2.1623802319717598,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004819427140697635,
      "loss": 5.5562,
      "mean_token_accuracy": 0.14681099206209183,
      "num_tokens": 4745070.0,
      "step": 2145
    },
    {
      "entropy": 5.947874927520752,
      "epoch": 2.167423096318709,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004817866951136753,
      "loss": 5.6263,
      "mean_token_accuracy": 0.14205105900764464,
      "num_tokens": 4756066.0,
      "step": 2150
    },
    {
      "entropy": 5.946091842651367,
      "epoch": 2.172465960665658,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048163003345976847,
      "loss": 5.6249,
      "mean_token_accuracy": 0.14056657999753952,
      "num_tokens": 4767153.0,
      "step": 2155
    },
    {
      "entropy": 5.883247137069702,
      "epoch": 2.177508825012607,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004814727295949512,
      "loss": 5.5406,
      "mean_token_accuracy": 0.14822641015052795,
      "num_tokens": 4777554.0,
      "step": 2160
    },
    {
      "entropy": 5.855099630355835,
      "epoch": 2.1825516893595562,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048131478400812735,
      "loss": 5.58,
      "mean_token_accuracy": 0.14793250113725662,
      "num_tokens": 4788881.0,
      "step": 2165
    },
    {
      "entropy": 5.950084686279297,
      "epoch": 2.1875945537065054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004811561971901956,
      "loss": 5.5617,
      "mean_token_accuracy": 0.14551590234041215,
      "num_tokens": 4798680.0,
      "step": 2170
    },
    {
      "entropy": 5.90962929725647,
      "epoch": 2.1926374180534545,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004809969696340474,
      "loss": 5.5424,
      "mean_token_accuracy": 0.14731819033622742,
      "num_tokens": 4808096.0,
      "step": 2175
    },
    {
      "entropy": 5.853575611114502,
      "epoch": 2.1976802824004036,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048083710183456554,
      "loss": 5.5016,
      "mean_token_accuracy": 0.15830671787261963,
      "num_tokens": 4818778.0,
      "step": 2180
    },
    {
      "entropy": 5.8463743209838865,
      "epoch": 2.2027231467473527,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048067659428862306,
      "loss": 5.5099,
      "mean_token_accuracy": 0.15291596204042435,
      "num_tokens": 4828586.0,
      "step": 2185
    },
    {
      "entropy": 5.879240798950195,
      "epoch": 2.2077660110943014,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048051544749508096,
      "loss": 5.5964,
      "mean_token_accuracy": 0.1415431432425976,
      "num_tokens": 4839419.0,
      "step": 2190
    },
    {
      "entropy": 5.8528117656707765,
      "epoch": 2.2128088754412505,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004803536619547872,
      "loss": 5.5253,
      "mean_token_accuracy": 0.1551996648311615,
      "num_tokens": 4849700.0,
      "step": 2195
    },
    {
      "entropy": 5.927838945388794,
      "epoch": 2.2178517397881996,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000480191238170575,
      "loss": 5.6865,
      "mean_token_accuracy": 0.14072581827640535,
      "num_tokens": 4860261.0,
      "step": 2200
    },
    {
      "entropy": 5.942123126983643,
      "epoch": 2.2228946041351487,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004800281766472612,
      "loss": 5.606,
      "mean_token_accuracy": 0.14752493649721146,
      "num_tokens": 4872277.0,
      "step": 2205
    },
    {
      "entropy": 5.9327787399292,
      "epoch": 2.227937468482098,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047986447789164486,
      "loss": 5.6514,
      "mean_token_accuracy": 0.1517902985215187,
      "num_tokens": 4883512.0,
      "step": 2210
    },
    {
      "entropy": 5.868672037124634,
      "epoch": 2.232980332829047,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004797001424125054,
      "loss": 5.5791,
      "mean_token_accuracy": 0.15303744301199912,
      "num_tokens": 4896197.0,
      "step": 2215
    },
    {
      "entropy": 5.887090826034546,
      "epoch": 2.238023197175996,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047953517072060124,
      "loss": 5.5484,
      "mean_token_accuracy": 0.15226840674877168,
      "num_tokens": 4907397.0,
      "step": 2220
    },
    {
      "entropy": 5.871148490905762,
      "epoch": 2.243066061522945,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004793695633286684,
      "loss": 5.6034,
      "mean_token_accuracy": 0.151021346449852,
      "num_tokens": 4919008.0,
      "step": 2225
    },
    {
      "entropy": 5.838630294799804,
      "epoch": 2.2481089258698943,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047920332075141825,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1499991774559021,
      "num_tokens": 4930877.0,
      "step": 2230
    },
    {
      "entropy": 5.875366067886352,
      "epoch": 2.253151790216843,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004790364435055367,
      "loss": 5.4496,
      "mean_token_accuracy": 0.1593221455812454,
      "num_tokens": 4942274.0,
      "step": 2235
    },
    {
      "entropy": 5.812536191940308,
      "epoch": 2.258194654563792,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000478868932109682,
      "loss": 5.5358,
      "mean_token_accuracy": 0.1527387946844101,
      "num_tokens": 4953740.0,
      "step": 2240
    },
    {
      "entropy": 5.902262592315674,
      "epoch": 2.263237518910741,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004787007870844835,
      "loss": 5.5712,
      "mean_token_accuracy": 0.1538744106888771,
      "num_tokens": 4964760.0,
      "step": 2245
    },
    {
      "entropy": 5.868859958648682,
      "epoch": 2.2682803832576903,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047853200895253976,
      "loss": 5.6361,
      "mean_token_accuracy": 0.14738022089004515,
      "num_tokens": 4976404.0,
      "step": 2250
    },
    {
      "entropy": 5.898247623443604,
      "epoch": 2.2733232476046394,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.517,
      "mean_token_accuracy": 0.14728200286626816,
      "num_tokens": 4987552.0,
      "step": 2255
    },
    {
      "entropy": 5.844000768661499,
      "epoch": 2.2783661119515886,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047819255546864797,
      "loss": 5.5286,
      "mean_token_accuracy": 0.14984559565782546,
      "num_tokens": 4998912.0,
      "step": 2260
    },
    {
      "entropy": 5.887818384170532,
      "epoch": 2.2834089762985377,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004780218811717293,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1581410750746727,
      "num_tokens": 5009556.0,
      "step": 2265
    },
    {
      "entropy": 5.834558296203613,
      "epoch": 2.288451840645487,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047785057587812057,
      "loss": 5.5505,
      "mean_token_accuracy": 0.15399980396032334,
      "num_tokens": 5021569.0,
      "step": 2270
    },
    {
      "entropy": 5.890166759490967,
      "epoch": 2.293494704992436,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047767864012024266,
      "loss": 5.6262,
      "mean_token_accuracy": 0.14727912992238998,
      "num_tokens": 5032994.0,
      "step": 2275
    },
    {
      "entropy": 5.945313930511475,
      "epoch": 2.2985375693393846,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047750607443247585,
      "loss": 5.5581,
      "mean_token_accuracy": 0.14870548546314238,
      "num_tokens": 5044681.0,
      "step": 2280
    },
    {
      "entropy": 5.793630743026734,
      "epoch": 2.3035804336863337,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047733287935115837,
      "loss": 5.4719,
      "mean_token_accuracy": 0.15815871804952622,
      "num_tokens": 5055509.0,
      "step": 2285
    },
    {
      "entropy": 5.76172251701355,
      "epoch": 2.308623298033283,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004771590554145844,
      "loss": 5.5002,
      "mean_token_accuracy": 0.16068245470523834,
      "num_tokens": 5066903.0,
      "step": 2290
    },
    {
      "entropy": 5.854470109939575,
      "epoch": 2.313666162380232,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047698460316300285,
      "loss": 5.5504,
      "mean_token_accuracy": 0.14702785909175872,
      "num_tokens": 5078411.0,
      "step": 2295
    },
    {
      "entropy": 5.862625408172607,
      "epoch": 2.318709026727181,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004768095231386153,
      "loss": 5.5565,
      "mean_token_accuracy": 0.14897067844867706,
      "num_tokens": 5089811.0,
      "step": 2300
    },
    {
      "entropy": 5.883305549621582,
      "epoch": 2.32375189107413,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004766338158855744,
      "loss": 5.5733,
      "mean_token_accuracy": 0.15708377063274384,
      "num_tokens": 5101196.0,
      "step": 2305
    },
    {
      "entropy": 5.832779026031494,
      "epoch": 2.3287947554210793,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004764574819499826,
      "loss": 5.5436,
      "mean_token_accuracy": 0.15225967913866043,
      "num_tokens": 5112379.0,
      "step": 2310
    },
    {
      "entropy": 5.895620441436767,
      "epoch": 2.3338376197680284,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047628052187988955,
      "loss": 5.6705,
      "mean_token_accuracy": 0.141206493973732,
      "num_tokens": 5125065.0,
      "step": 2315
    },
    {
      "entropy": 5.997529363632202,
      "epoch": 2.338880484114977,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004761029362252914,
      "loss": 5.6291,
      "mean_token_accuracy": 0.14648888558149337,
      "num_tokens": 5137471.0,
      "step": 2320
    },
    {
      "entropy": 5.9020641326904295,
      "epoch": 2.343923348461926,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047592472553812856,
      "loss": 5.5518,
      "mean_token_accuracy": 0.15038047060370446,
      "num_tokens": 5147682.0,
      "step": 2325
    },
    {
      "entropy": 5.817275094985962,
      "epoch": 2.3489662128088753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004757458903722837,
      "loss": 5.5417,
      "mean_token_accuracy": 0.14693662226200105,
      "num_tokens": 5156143.0,
      "step": 2330
    },
    {
      "entropy": 5.934580278396607,
      "epoch": 2.3540090771558244,
      "grad_norm": 0.875,
      "learning_rate": 0.00047556643128358075,
      "loss": 5.5985,
      "mean_token_accuracy": 0.14841473698616028,
      "num_tokens": 5167462.0,
      "step": 2335
    },
    {
      "entropy": 5.744073915481567,
      "epoch": 2.3590519415027735,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047538634882978277,
      "loss": 5.5169,
      "mean_token_accuracy": 0.15127300173044206,
      "num_tokens": 5178030.0,
      "step": 2340
    },
    {
      "entropy": 5.785821199417114,
      "epoch": 2.3640948058497226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047520564357059013,
      "loss": 5.5364,
      "mean_token_accuracy": 0.15155871361494064,
      "num_tokens": 5188296.0,
      "step": 2345
    },
    {
      "entropy": 5.808376026153565,
      "epoch": 2.3691376701966718,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004750243160676389,
      "loss": 5.5265,
      "mean_token_accuracy": 0.1540423169732094,
      "num_tokens": 5199058.0,
      "step": 2350
    },
    {
      "entropy": 5.812775659561157,
      "epoch": 2.374180534543621,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004748423668844992,
      "loss": 5.498,
      "mean_token_accuracy": 0.15063866078853608,
      "num_tokens": 5209314.0,
      "step": 2355
    },
    {
      "entropy": 5.90487060546875,
      "epoch": 2.37922339889057,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004746597965866733,
      "loss": 5.6177,
      "mean_token_accuracy": 0.14394214525818824,
      "num_tokens": 5220248.0,
      "step": 2360
    },
    {
      "entropy": 5.914368104934693,
      "epoch": 2.384266263237519,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004744766057415938,
      "loss": 5.5947,
      "mean_token_accuracy": 0.1400570899248123,
      "num_tokens": 5230874.0,
      "step": 2365
    },
    {
      "entropy": 5.826127243041992,
      "epoch": 2.389309127584468,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004742927949186222,
      "loss": 5.5239,
      "mean_token_accuracy": 0.15305102318525315,
      "num_tokens": 5243328.0,
      "step": 2370
    },
    {
      "entropy": 5.763830518722534,
      "epoch": 2.394351991931417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047410836468904664,
      "loss": 5.4773,
      "mean_token_accuracy": 0.14843705296516418,
      "num_tokens": 5254357.0,
      "step": 2375
    },
    {
      "entropy": 5.795266103744507,
      "epoch": 2.399394856278366,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047392331562608073,
      "loss": 5.4802,
      "mean_token_accuracy": 0.15369822680950165,
      "num_tokens": 5264492.0,
      "step": 2380
    },
    {
      "entropy": 5.744703674316407,
      "epoch": 2.404437720625315,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004737376483048609,
      "loss": 5.443,
      "mean_token_accuracy": 0.1589686468243599,
      "num_tokens": 5274816.0,
      "step": 2385
    },
    {
      "entropy": 5.8192870140075685,
      "epoch": 2.4094805849722642,
      "grad_norm": 1.0,
      "learning_rate": 0.0004735513633024458,
      "loss": 5.5429,
      "mean_token_accuracy": 0.14981934279203415,
      "num_tokens": 5286855.0,
      "step": 2390
    },
    {
      "entropy": 5.903432035446167,
      "epoch": 2.4145234493192134,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004733644611978134,
      "loss": 5.6537,
      "mean_token_accuracy": 0.13768935427069665,
      "num_tokens": 5299594.0,
      "step": 2395
    },
    {
      "entropy": 5.8120519638061525,
      "epoch": 2.4195663136661625,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004731769425718597,
      "loss": 5.4793,
      "mean_token_accuracy": 0.15393216013908387,
      "num_tokens": 5310078.0,
      "step": 2400
    },
    {
      "entropy": 5.746055555343628,
      "epoch": 2.4246091780131116,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000472988808007397,
      "loss": 5.5048,
      "mean_token_accuracy": 0.15920527279376984,
      "num_tokens": 5319544.0,
      "step": 2405
    },
    {
      "entropy": 5.794832992553711,
      "epoch": 2.4296520423600603,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00047280005808915176,
      "loss": 5.568,
      "mean_token_accuracy": 0.14523061513900756,
      "num_tokens": 5331025.0,
      "step": 2410
    },
    {
      "entropy": 5.790472984313965,
      "epoch": 2.4346949067070094,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004726106934037631,
      "loss": 5.5414,
      "mean_token_accuracy": 0.15091371536254883,
      "num_tokens": 5343086.0,
      "step": 2415
    },
    {
      "entropy": 5.923413705825806,
      "epoch": 2.4397377710539585,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.5993,
      "mean_token_accuracy": 0.15322286635637283,
      "num_tokens": 5353598.0,
      "step": 2420
    },
    {
      "entropy": 5.803432273864746,
      "epoch": 2.4447806354009076,
      "grad_norm": 1.125,
      "learning_rate": 0.00047223012208766384,
      "loss": 5.4755,
      "mean_token_accuracy": 0.14711593165993692,
      "num_tokens": 5363607.0,
      "step": 2425
    },
    {
      "entropy": 5.779847431182861,
      "epoch": 2.4498234997478567,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004720389166397777,
      "loss": 5.5388,
      "mean_token_accuracy": 0.15192387402057647,
      "num_tokens": 5375198.0,
      "step": 2430
    },
    {
      "entropy": 5.7408387660980225,
      "epoch": 2.454866364094806,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047184709879039325,
      "loss": 5.5235,
      "mean_token_accuracy": 0.149568809568882,
      "num_tokens": 5386058.0,
      "step": 2435
    },
    {
      "entropy": 5.9129396915435795,
      "epoch": 2.459909228441755,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047165466913568497,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1510582074522972,
      "num_tokens": 5395720.0,
      "step": 2440
    },
    {
      "entropy": 5.781081771850586,
      "epoch": 2.464952092788704,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004714616282737285,
      "loss": 5.5995,
      "mean_token_accuracy": 0.14857714772224426,
      "num_tokens": 5406252.0,
      "step": 2445
    },
    {
      "entropy": 5.839043188095093,
      "epoch": 2.469994957135653,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004712679768044995,
      "loss": 5.603,
      "mean_token_accuracy": 0.14704830050468445,
      "num_tokens": 5418125.0,
      "step": 2450
    },
    {
      "entropy": 5.849780893325805,
      "epoch": 2.4750378214826023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000471073715329871,
      "loss": 5.5515,
      "mean_token_accuracy": 0.15433166921138763,
      "num_tokens": 5428095.0,
      "step": 2455
    },
    {
      "entropy": 5.860912895202636,
      "epoch": 2.4800806858295514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004708788444536123,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1464594319462776,
      "num_tokens": 5438359.0,
      "step": 2460
    },
    {
      "entropy": 5.805293464660645,
      "epoch": 2.4851235501765,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004706833647813865,
      "loss": 5.4992,
      "mean_token_accuracy": 0.15880793184041977,
      "num_tokens": 5448559.0,
      "step": 2465
    },
    {
      "entropy": 5.751124143600464,
      "epoch": 2.4901664145234492,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047048727692074893,
      "loss": 5.5168,
      "mean_token_accuracy": 0.14838636070489883,
      "num_tokens": 5459739.0,
      "step": 2470
    },
    {
      "entropy": 5.7891887664794925,
      "epoch": 2.4952092788703983,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004702905814811452,
      "loss": 5.4923,
      "mean_token_accuracy": 0.14752392023801802,
      "num_tokens": 5471160.0,
      "step": 2475
    },
    {
      "entropy": 5.707677030563355,
      "epoch": 2.5002521432173475,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00047009327907390926,
      "loss": 5.5503,
      "mean_token_accuracy": 0.14980040788650512,
      "num_tokens": 5482344.0,
      "step": 2480
    },
    {
      "entropy": 5.856988191604614,
      "epoch": 2.5052950075642966,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004698953703122616,
      "loss": 5.5528,
      "mean_token_accuracy": 0.1503003254532814,
      "num_tokens": 5493111.0,
      "step": 2485
    },
    {
      "entropy": 5.655979204177856,
      "epoch": 2.5103378719112457,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046969685581130715,
      "loss": 5.4873,
      "mean_token_accuracy": 0.1560885176062584,
      "num_tokens": 5503877.0,
      "step": 2490
    },
    {
      "entropy": 5.831366395950317,
      "epoch": 2.515380736258195,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046949773618803353,
      "loss": 5.5051,
      "mean_token_accuracy": 0.15249394997954369,
      "num_tokens": 5514067.0,
      "step": 2495
    },
    {
      "entropy": 5.860077047348023,
      "epoch": 2.5204236006051435,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00046929801206130925,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1510472685098648,
      "num_tokens": 5525534.0,
      "step": 2500
    },
    {
      "epoch": 2.5204236006051435,
      "eval_entropy": 5.669652560285029,
      "eval_loss": 5.856481552124023,
      "eval_mean_token_accuracy": 0.14118836258962258,
      "eval_num_tokens": 5525534.0,
      "eval_runtime": 2.0404,
      "eval_samples_per_second": 1537.926,
      "eval_steps_per_second": 192.608,
      "step": 2500
    },
    {
      "entropy": 5.819248008728027,
      "epoch": 2.5254664649520926,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046909768405188135,
      "loss": 5.5295,
      "mean_token_accuracy": 0.15700920820236205,
      "num_tokens": 5535525.0,
      "step": 2505
    },
    {
      "entropy": 5.769597578048706,
      "epoch": 2.5305093292990417,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046889675278237404,
      "loss": 5.52,
      "mean_token_accuracy": 0.15728203654289247,
      "num_tokens": 5546129.0,
      "step": 2510
    },
    {
      "entropy": 5.912272644042969,
      "epoch": 2.535552193645991,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004686952188772863,
      "loss": 5.6652,
      "mean_token_accuracy": 0.14891352951526643,
      "num_tokens": 5557548.0,
      "step": 2515
    },
    {
      "entropy": 5.8772557258605955,
      "epoch": 2.54059505799294,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046849308296299005,
      "loss": 5.5734,
      "mean_token_accuracy": 0.15616171360015868,
      "num_tokens": 5567367.0,
      "step": 2520
    },
    {
      "entropy": 5.810514688491821,
      "epoch": 2.545637922339889,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004682903456677285,
      "loss": 5.4811,
      "mean_token_accuracy": 0.15235700085759163,
      "num_tokens": 5577489.0,
      "step": 2525
    },
    {
      "entropy": 5.77357177734375,
      "epoch": 2.550680786686838,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046808700762161385,
      "loss": 5.5288,
      "mean_token_accuracy": 0.1467547282576561,
      "num_tokens": 5587877.0,
      "step": 2530
    },
    {
      "entropy": 5.770196580886841,
      "epoch": 2.5557236510337873,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004678830694566253,
      "loss": 5.5394,
      "mean_token_accuracy": 0.15670244991779328,
      "num_tokens": 5597863.0,
      "step": 2535
    },
    {
      "entropy": 5.807074022293091,
      "epoch": 2.5607665153807364,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004676785318066075,
      "loss": 5.5454,
      "mean_token_accuracy": 0.153423710167408,
      "num_tokens": 5608141.0,
      "step": 2540
    },
    {
      "entropy": 5.787803220748901,
      "epoch": 2.5658093797276855,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004674733953072682,
      "loss": 5.6115,
      "mean_token_accuracy": 0.14963685870170593,
      "num_tokens": 5619660.0,
      "step": 2545
    },
    {
      "entropy": 5.947618913650513,
      "epoch": 2.5708522440746346,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004672676605961763,
      "loss": 5.6018,
      "mean_token_accuracy": 0.14351876080036163,
      "num_tokens": 5630271.0,
      "step": 2550
    },
    {
      "entropy": 5.773510217666626,
      "epoch": 2.5758951084215833,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046706132831276007,
      "loss": 5.587,
      "mean_token_accuracy": 0.15054969489574432,
      "num_tokens": 5642594.0,
      "step": 2555
    },
    {
      "entropy": 5.778890609741211,
      "epoch": 2.5809379727685324,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046685439909830486,
      "loss": 5.5241,
      "mean_token_accuracy": 0.15210181772708892,
      "num_tokens": 5653361.0,
      "step": 2560
    },
    {
      "entropy": 5.767368650436401,
      "epoch": 2.5859808371154815,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004666468735959516,
      "loss": 5.438,
      "mean_token_accuracy": 0.1562154620885849,
      "num_tokens": 5664167.0,
      "step": 2565
    },
    {
      "entropy": 5.761620044708252,
      "epoch": 2.5910237014624307,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664387524506943,
      "loss": 5.5465,
      "mean_token_accuracy": 0.14990140795707702,
      "num_tokens": 5675903.0,
      "step": 2570
    },
    {
      "entropy": 5.845316219329834,
      "epoch": 2.59606656580938,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004662300363093784,
      "loss": 5.522,
      "mean_token_accuracy": 0.15081020295619965,
      "num_tokens": 5686319.0,
      "step": 2575
    },
    {
      "entropy": 5.9148040294647215,
      "epoch": 2.601109430156329,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004660207258206983,
      "loss": 5.6162,
      "mean_token_accuracy": 0.14688173905014992,
      "num_tokens": 5696000.0,
      "step": 2580
    },
    {
      "entropy": 5.7588094711303714,
      "epoch": 2.606152294503278,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.5167,
      "mean_token_accuracy": 0.15163008272647857,
      "num_tokens": 5708145.0,
      "step": 2585
    },
    {
      "entropy": 5.725979423522949,
      "epoch": 2.6111951588502267,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004656003244052582,
      "loss": 5.4874,
      "mean_token_accuracy": 0.1598169982433319,
      "num_tokens": 5718053.0,
      "step": 2590
    },
    {
      "entropy": 5.767901945114136,
      "epoch": 2.616238023197176,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004653892347851155,
      "loss": 5.5493,
      "mean_token_accuracy": 0.15450764447450638,
      "num_tokens": 5728723.0,
      "step": 2595
    },
    {
      "entropy": 5.755375957489013,
      "epoch": 2.621280887544125,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004651775534308392,
      "loss": 5.475,
      "mean_token_accuracy": 0.16065322309732438,
      "num_tokens": 5739414.0,
      "step": 2600
    },
    {
      "entropy": 5.794290542602539,
      "epoch": 2.626323751891074,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004649652810003398,
      "loss": 5.625,
      "mean_token_accuracy": 0.14363892748951912,
      "num_tokens": 5751247.0,
      "step": 2605
    },
    {
      "entropy": 5.791951465606689,
      "epoch": 2.631366616238023,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046475241815336504,
      "loss": 5.4823,
      "mean_token_accuracy": 0.15644117295742035,
      "num_tokens": 5761791.0,
      "step": 2610
    },
    {
      "entropy": 5.740055847167969,
      "epoch": 2.6364094805849723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645389655514974,
      "loss": 5.4744,
      "mean_token_accuracy": 0.14935198575258254,
      "num_tokens": 5773597.0,
      "step": 2615
    },
    {
      "entropy": 5.794382095336914,
      "epoch": 2.6414523449319214,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004643249238581525,
      "loss": 5.4981,
      "mean_token_accuracy": 0.15026383846998215,
      "num_tokens": 5782963.0,
      "step": 2620
    },
    {
      "entropy": 5.6868040561676025,
      "epoch": 2.6464952092788705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004641102937385769,
      "loss": 5.5234,
      "mean_token_accuracy": 0.14880652129650115,
      "num_tokens": 5793474.0,
      "step": 2625
    },
    {
      "entropy": 5.7942564487457275,
      "epoch": 2.6515380736258196,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004638950758598461,
      "loss": 5.5086,
      "mean_token_accuracy": 0.1518484354019165,
      "num_tokens": 5804018.0,
      "step": 2630
    },
    {
      "entropy": 5.857076978683471,
      "epoch": 2.6565809379727687,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046367927089086203,
      "loss": 5.5453,
      "mean_token_accuracy": 0.14937240481376649,
      "num_tokens": 5815897.0,
      "step": 2635
    },
    {
      "entropy": 5.783289623260498,
      "epoch": 2.661623802319718,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046346287950235163,
      "loss": 5.5718,
      "mean_token_accuracy": 0.14845054149627684,
      "num_tokens": 5825883.0,
      "step": 2640
    },
    {
      "entropy": 5.835809564590454,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004632459023668642,
      "loss": 5.534,
      "mean_token_accuracy": 0.15753208100795746,
      "num_tokens": 5836668.0,
      "step": 2645
    },
    {
      "entropy": 5.812521171569824,
      "epoch": 2.6717095310136156,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004630283401587699,
      "loss": 5.6635,
      "mean_token_accuracy": 0.14314896017313003,
      "num_tokens": 5848497.0,
      "step": 2650
    },
    {
      "entropy": 5.7993275165557865,
      "epoch": 2.6767523953605648,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004628101935542568,
      "loss": 5.476,
      "mean_token_accuracy": 0.15989356487989426,
      "num_tokens": 5859231.0,
      "step": 2655
    },
    {
      "entropy": 5.795991325378418,
      "epoch": 2.681795259707514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046259146323132986,
      "loss": 5.6306,
      "mean_token_accuracy": 0.14957452565431595,
      "num_tokens": 5871190.0,
      "step": 2660
    },
    {
      "entropy": 5.786652040481568,
      "epoch": 2.686838124054463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046237214986980776,
      "loss": 5.4843,
      "mean_token_accuracy": 0.15117899924516678,
      "num_tokens": 5881788.0,
      "step": 2665
    },
    {
      "entropy": 5.762012195587158,
      "epoch": 2.691880988401412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004621522541513216,
      "loss": 5.4693,
      "mean_token_accuracy": 0.1548840656876564,
      "num_tokens": 5891289.0,
      "step": 2670
    },
    {
      "entropy": 5.71791353225708,
      "epoch": 2.696923852748361,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004619317767593123,
      "loss": 5.5099,
      "mean_token_accuracy": 0.15003726333379747,
      "num_tokens": 5903594.0,
      "step": 2675
    },
    {
      "entropy": 5.874940586090088,
      "epoch": 2.70196671709531,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046171071837902873,
      "loss": 5.6303,
      "mean_token_accuracy": 0.15058379024267196,
      "num_tokens": 5914592.0,
      "step": 2680
    },
    {
      "entropy": 5.747613334655762,
      "epoch": 2.707009581442259,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004614890796975254,
      "loss": 5.4503,
      "mean_token_accuracy": 0.15491787791252137,
      "num_tokens": 5927137.0,
      "step": 2685
    },
    {
      "entropy": 5.760874032974243,
      "epoch": 2.712052445789208,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004612668614036605,
      "loss": 5.5606,
      "mean_token_accuracy": 0.15194035768508912,
      "num_tokens": 5938730.0,
      "step": 2690
    },
    {
      "entropy": 5.808653068542481,
      "epoch": 2.7170953101361572,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004610440641880934,
      "loss": 5.5755,
      "mean_token_accuracy": 0.14960916787385942,
      "num_tokens": 5948912.0,
      "step": 2695
    },
    {
      "entropy": 5.733116531372071,
      "epoch": 2.7221381744831064,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046082068874328324,
      "loss": 5.5035,
      "mean_token_accuracy": 0.1540943205356598,
      "num_tokens": 5960253.0,
      "step": 2700
    },
    {
      "entropy": 5.814432811737061,
      "epoch": 2.7271810388300555,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046059673576348596,
      "loss": 5.5983,
      "mean_token_accuracy": 0.14576624780893327,
      "num_tokens": 5972615.0,
      "step": 2705
    },
    {
      "entropy": 5.867447233200073,
      "epoch": 2.7322239031770046,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046037220594475264,
      "loss": 5.5646,
      "mean_token_accuracy": 0.1537248432636261,
      "num_tokens": 5985023.0,
      "step": 2710
    },
    {
      "entropy": 5.686407423019409,
      "epoch": 2.7372667675239537,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000460147099984927,
      "loss": 5.4651,
      "mean_token_accuracy": 0.1570134088397026,
      "num_tokens": 5996445.0,
      "step": 2715
    },
    {
      "entropy": 5.789271497726441,
      "epoch": 2.742309631870903,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004599214185836438,
      "loss": 5.5116,
      "mean_token_accuracy": 0.14551102817058564,
      "num_tokens": 6007385.0,
      "step": 2720
    },
    {
      "entropy": 5.780018663406372,
      "epoch": 2.747352496217852,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045969516244232574,
      "loss": 5.5484,
      "mean_token_accuracy": 0.15154093354940415,
      "num_tokens": 6018174.0,
      "step": 2725
    },
    {
      "entropy": 5.726502370834351,
      "epoch": 2.752395360564801,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045946833226418235,
      "loss": 5.469,
      "mean_token_accuracy": 0.14863964468240737,
      "num_tokens": 6030049.0,
      "step": 2730
    },
    {
      "entropy": 5.716791200637817,
      "epoch": 2.7574382249117497,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004592409287542069,
      "loss": 5.5229,
      "mean_token_accuracy": 0.15207778811454772,
      "num_tokens": 6040155.0,
      "step": 2735
    },
    {
      "entropy": 5.703477764129639,
      "epoch": 2.762481089258699,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004590129526191748,
      "loss": 5.5284,
      "mean_token_accuracy": 0.15693228095769882,
      "num_tokens": 6050569.0,
      "step": 2740
    },
    {
      "entropy": 5.795703887939453,
      "epoch": 2.767523953605648,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004587844045676413,
      "loss": 5.4791,
      "mean_token_accuracy": 0.15692086666822433,
      "num_tokens": 6060871.0,
      "step": 2745
    },
    {
      "entropy": 5.748446369171143,
      "epoch": 2.772566817952597,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.4841,
      "mean_token_accuracy": 0.16212048679590224,
      "num_tokens": 6073045.0,
      "step": 2750
    },
    {
      "entropy": 5.743357181549072,
      "epoch": 2.777609682299546,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045832559555817525,
      "loss": 5.5294,
      "mean_token_accuracy": 0.15036371648311614,
      "num_tokens": 6084798.0,
      "step": 2755
    },
    {
      "entropy": 5.762336921691895,
      "epoch": 2.7826525466464953,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045809533602623185,
      "loss": 5.4675,
      "mean_token_accuracy": 0.15212155729532242,
      "num_tokens": 6095831.0,
      "step": 2760
    },
    {
      "entropy": 5.75358247756958,
      "epoch": 2.7876954109934444,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004578645074297603,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1555684432387352,
      "num_tokens": 6106958.0,
      "step": 2765
    },
    {
      "entropy": 5.71989336013794,
      "epoch": 2.792738275340393,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045763311048618144,
      "loss": 5.473,
      "mean_token_accuracy": 0.1608845993876457,
      "num_tokens": 6118165.0,
      "step": 2770
    },
    {
      "entropy": 5.665394067764282,
      "epoch": 2.797781139687342,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004574011459146822,
      "loss": 5.4641,
      "mean_token_accuracy": 0.1566280245780945,
      "num_tokens": 6129297.0,
      "step": 2775
    },
    {
      "entropy": 5.766199827194214,
      "epoch": 2.8028240040342913,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00045716861443621386,
      "loss": 5.4923,
      "mean_token_accuracy": 0.15070500820875168,
      "num_tokens": 6140071.0,
      "step": 2780
    },
    {
      "entropy": 5.747651100158691,
      "epoch": 2.8078668683812404,
      "grad_norm": 0.9375,
      "learning_rate": 0.00045693551677348967,
      "loss": 5.4975,
      "mean_token_accuracy": 0.15279183983802797,
      "num_tokens": 6150940.0,
      "step": 2785
    },
    {
      "entropy": 5.80690107345581,
      "epoch": 2.8129097327281896,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004567018536509823,
      "loss": 5.4803,
      "mean_token_accuracy": 0.15254825353622437,
      "num_tokens": 6161966.0,
      "step": 2790
    },
    {
      "entropy": 5.824843215942383,
      "epoch": 2.8179525970751387,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00045646762579492245,
      "loss": 5.5348,
      "mean_token_accuracy": 0.15181106626987456,
      "num_tokens": 6174455.0,
      "step": 2795
    },
    {
      "entropy": 5.693896532058716,
      "epoch": 2.822995461422088,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045623283393329546,
      "loss": 5.4716,
      "mean_token_accuracy": 0.15222429782152175,
      "num_tokens": 6184789.0,
      "step": 2800
    },
    {
      "entropy": 5.757958793640137,
      "epoch": 2.828038325769037,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045599747879584,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1462457463145256,
      "num_tokens": 6195704.0,
      "step": 2805
    },
    {
      "entropy": 5.788491535186767,
      "epoch": 2.833081190115986,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004557615611140451,
      "loss": 5.4409,
      "mean_token_accuracy": 0.15523429661989213,
      "num_tokens": 6206290.0,
      "step": 2810
    },
    {
      "entropy": 5.7461964130401615,
      "epoch": 2.838124054462935,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004555250816211485,
      "loss": 5.5513,
      "mean_token_accuracy": 0.1454817198216915,
      "num_tokens": 6216660.0,
      "step": 2815
    },
    {
      "entropy": 5.743470144271851,
      "epoch": 2.8431669188098843,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00045528804105213395,
      "loss": 5.4974,
      "mean_token_accuracy": 0.15376585721969604,
      "num_tokens": 6228005.0,
      "step": 2820
    },
    {
      "entropy": 5.701382303237915,
      "epoch": 2.848209783156833,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004550504401437289,
      "loss": 5.452,
      "mean_token_accuracy": 0.15123870223760605,
      "num_tokens": 6239447.0,
      "step": 2825
    },
    {
      "entropy": 5.765235710144043,
      "epoch": 2.853252647503782,
      "grad_norm": 0.84375,
      "learning_rate": 0.00045481227963440235,
      "loss": 5.5613,
      "mean_token_accuracy": 0.14397587329149247,
      "num_tokens": 6252312.0,
      "step": 2830
    },
    {
      "entropy": 5.774335527420044,
      "epoch": 2.858295511850731,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045457356026436295,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1525854289531708,
      "num_tokens": 6264131.0,
      "step": 2835
    },
    {
      "entropy": 5.689094018936157,
      "epoch": 2.8633383761976803,
      "grad_norm": 0.890625,
      "learning_rate": 0.00045433428277555577,
      "loss": 5.4477,
      "mean_token_accuracy": 0.15649830549955368,
      "num_tokens": 6275161.0,
      "step": 2840
    },
    {
      "entropy": 5.7043664932250975,
      "epoch": 2.8683812405446294,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004540944479116609,
      "loss": 5.4985,
      "mean_token_accuracy": 0.1511241763830185,
      "num_tokens": 6285655.0,
      "step": 2845
    },
    {
      "entropy": 5.798206949234009,
      "epoch": 2.8734241048915785,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004538540564180906,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15425564870238304,
      "num_tokens": 6296532.0,
      "step": 2850
    },
    {
      "entropy": 5.878371000289917,
      "epoch": 2.8784669692385276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004536131090419871,
      "loss": 5.6173,
      "mean_token_accuracy": 0.14916557520627977,
      "num_tokens": 6308229.0,
      "step": 2855
    },
    {
      "entropy": 5.743052625656128,
      "epoch": 2.8835098335854763,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004533716065322205,
      "loss": 5.5368,
      "mean_token_accuracy": 0.15372492074966432,
      "num_tokens": 6321257.0,
      "step": 2860
    },
    {
      "entropy": 5.792555570602417,
      "epoch": 2.8885526979324254,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004531295496393861,
      "loss": 5.492,
      "mean_token_accuracy": 0.15237339586019516,
      "num_tokens": 6332040.0,
      "step": 2865
    },
    {
      "entropy": 5.679398250579834,
      "epoch": 2.8935955622793745,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004528869391158023,
      "loss": 5.3587,
      "mean_token_accuracy": 0.16076318472623824,
      "num_tokens": 6342925.0,
      "step": 2870
    },
    {
      "entropy": 5.777550983428955,
      "epoch": 2.8986384266263236,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00045264377571550834,
      "loss": 5.5478,
      "mean_token_accuracy": 0.15306758880615234,
      "num_tokens": 6354263.0,
      "step": 2875
    },
    {
      "entropy": 5.823110675811767,
      "epoch": 2.9036812909732728,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004524000601942615,
      "loss": 5.5441,
      "mean_token_accuracy": 0.14977220743894576,
      "num_tokens": 6365497.0,
      "step": 2880
    },
    {
      "entropy": 5.803255605697632,
      "epoch": 2.908724155320222,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004521557933095354,
      "loss": 5.594,
      "mean_token_accuracy": 0.1505901463329792,
      "num_tokens": 6376501.0,
      "step": 2885
    },
    {
      "entropy": 5.776741027832031,
      "epoch": 2.913767019667171,
      "grad_norm": 0.953125,
      "learning_rate": 0.000451910975820517,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1469830572605133,
      "num_tokens": 6387632.0,
      "step": 2890
    },
    {
      "entropy": 5.669276332855224,
      "epoch": 2.91880988401412,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045166560848810476,
      "loss": 5.5222,
      "mean_token_accuracy": 0.16028138548135756,
      "num_tokens": 6398813.0,
      "step": 2895
    },
    {
      "entropy": 5.841599655151367,
      "epoch": 2.9238527483610692,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045141969207490606,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1571269378066063,
      "num_tokens": 6409532.0,
      "step": 2900
    },
    {
      "entropy": 5.751657485961914,
      "epoch": 2.9288956127080183,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004511732273452347,
      "loss": 5.5746,
      "mean_token_accuracy": 0.14663779512047767,
      "num_tokens": 6420755.0,
      "step": 2905
    },
    {
      "entropy": 5.788761901855469,
      "epoch": 2.9339384770549675,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004509262150651089,
      "loss": 5.6045,
      "mean_token_accuracy": 0.15201486200094222,
      "num_tokens": 6430662.0,
      "step": 2910
    },
    {
      "entropy": 5.798057413101196,
      "epoch": 2.938981341401916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.4817,
      "mean_token_accuracy": 0.1538977727293968,
      "num_tokens": 6441230.0,
      "step": 2915
    },
    {
      "entropy": 5.799156188964844,
      "epoch": 2.9440242057488653,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004504305509260724,
      "loss": 5.6031,
      "mean_token_accuracy": 0.14440229386091233,
      "num_tokens": 6452574.0,
      "step": 2920
    },
    {
      "entropy": 5.729610824584961,
      "epoch": 2.9490670700958144,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004501819006076974,
      "loss": 5.4756,
      "mean_token_accuracy": 0.15835335105657578,
      "num_tokens": 6462562.0,
      "step": 2925
    },
    {
      "entropy": 5.711238670349121,
      "epoch": 2.9541099344427635,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044993270581993413,
      "loss": 5.4884,
      "mean_token_accuracy": 0.15400700718164445,
      "num_tokens": 6473594.0,
      "step": 2930
    },
    {
      "entropy": 5.761794567108154,
      "epoch": 2.9591527987897126,
      "grad_norm": 0.90625,
      "learning_rate": 0.00044968296733728587,
      "loss": 5.5127,
      "mean_token_accuracy": 0.15556546151638032,
      "num_tokens": 6484129.0,
      "step": 2935
    },
    {
      "entropy": 5.786276531219483,
      "epoch": 2.9641956631366617,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004494326859359456,
      "loss": 5.4905,
      "mean_token_accuracy": 0.14901109039783478,
      "num_tokens": 6495584.0,
      "step": 2940
    },
    {
      "entropy": 5.75063648223877,
      "epoch": 2.969238527483611,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044918186239379363,
      "loss": 5.5692,
      "mean_token_accuracy": 0.1525542750954628,
      "num_tokens": 6507872.0,
      "step": 2945
    },
    {
      "entropy": 5.712465095520019,
      "epoch": 2.9742813918305595,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004489304974903953,
      "loss": 5.4332,
      "mean_token_accuracy": 0.15936507284641266,
      "num_tokens": 6519172.0,
      "step": 2950
    },
    {
      "entropy": 5.729972171783447,
      "epoch": 2.9793242561775086,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004486785920069985,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15972347259521485,
      "num_tokens": 6530151.0,
      "step": 2955
    },
    {
      "entropy": 5.8090736865997314,
      "epoch": 2.9843671205244577,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004484261467265313,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1548163615167141,
      "num_tokens": 6541881.0,
      "step": 2960
    },
    {
      "entropy": 5.719935655593872,
      "epoch": 2.989409984871407,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00044817316243359956,
      "loss": 5.5406,
      "mean_token_accuracy": 0.15042774602770806,
      "num_tokens": 6553390.0,
      "step": 2965
    },
    {
      "entropy": 5.774695491790771,
      "epoch": 2.994452849218356,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044791963991448427,
      "loss": 5.3679,
      "mean_token_accuracy": 0.1631093814969063,
      "num_tokens": 6563754.0,
      "step": 2970
    },
    {
      "entropy": 5.682390308380127,
      "epoch": 2.999495713565305,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00044766557995713915,
      "loss": 5.4578,
      "mean_token_accuracy": 0.1516885504126549,
      "num_tokens": 6574124.0,
      "step": 2975
    },
    {
      "entropy": 5.759289847479926,
      "epoch": 3.0040342914775593,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004474109833511885,
      "loss": 5.2792,
      "mean_token_accuracy": 0.1634783943494161,
      "num_tokens": 6582795.0,
      "step": 2980
    },
    {
      "entropy": 5.714936923980713,
      "epoch": 3.0090771558245084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004471558508879244,
      "loss": 5.2117,
      "mean_token_accuracy": 0.1708674907684326,
      "num_tokens": 6593649.0,
      "step": 2985
    },
    {
      "entropy": 5.6874351978302,
      "epoch": 3.0141200201714575,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004469001833603045,
      "loss": 5.2149,
      "mean_token_accuracy": 0.17300137430429458,
      "num_tokens": 6606370.0,
      "step": 2990
    },
    {
      "entropy": 5.653540849685669,
      "epoch": 3.0191628845184066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004466439815629493,
      "loss": 5.1578,
      "mean_token_accuracy": 0.17079648822546006,
      "num_tokens": 6616599.0,
      "step": 2995
    },
    {
      "entropy": 5.7674613952636715,
      "epoch": 3.0242057488653553,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044638724629214,
      "loss": 5.2355,
      "mean_token_accuracy": 0.16412699967622757,
      "num_tokens": 6627707.0,
      "step": 3000
    },
    {
      "epoch": 3.0242057488653553,
      "eval_entropy": 5.550955645000662,
      "eval_loss": 5.774935245513916,
      "eval_mean_token_accuracy": 0.14621713238059716,
      "eval_num_tokens": 6627707.0,
      "eval_runtime": 2.0361,
      "eval_samples_per_second": 1541.159,
      "eval_steps_per_second": 193.013,
      "step": 3000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9910,
  "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": 1640980747284480.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}