ind-latn-10mb-after-ppt-Dp-100mb-ckpt500_seed455/checkpoint-12000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9336704921221552,
  "eval_steps": 3000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.34880576133728,
      "epoch": 0.00038902937171756465,
      "grad_norm": 1.0703125,
      "learning_rate": 2e-06,
      "loss": 7.0252,
      "mean_token_accuracy": 0.10256106480956077,
      "num_tokens": 9243.0,
      "step": 5
    },
    {
      "entropy": 7.446655797958374,
      "epoch": 0.0007780587434351293,
      "grad_norm": 1.1328125,
      "learning_rate": 4.5e-06,
      "loss": 7.1577,
      "mean_token_accuracy": 0.0929850086569786,
      "num_tokens": 18424.0,
      "step": 10
    },
    {
      "entropy": 7.352576017379761,
      "epoch": 0.001167088115152694,
      "grad_norm": 1.0703125,
      "learning_rate": 7e-06,
      "loss": 7.0451,
      "mean_token_accuracy": 0.0964082270860672,
      "num_tokens": 27006.0,
      "step": 15
    },
    {
      "entropy": 7.343383884429931,
      "epoch": 0.0015561174868702586,
      "grad_norm": 1.078125,
      "learning_rate": 9.5e-06,
      "loss": 7.0864,
      "mean_token_accuracy": 0.0972217358648777,
      "num_tokens": 35577.0,
      "step": 20
    },
    {
      "entropy": 7.42727689743042,
      "epoch": 0.0019451468585878235,
      "grad_norm": 0.96875,
      "learning_rate": 1.2e-05,
      "loss": 7.0596,
      "mean_token_accuracy": 0.09545032754540443,
      "num_tokens": 43706.0,
      "step": 25
    },
    {
      "entropy": 7.413392829895019,
      "epoch": 0.002334176230305388,
      "grad_norm": 1.0859375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.0333,
      "mean_token_accuracy": 0.09936482384800911,
      "num_tokens": 52571.0,
      "step": 30
    },
    {
      "entropy": 7.465293312072754,
      "epoch": 0.0027232056020229526,
      "grad_norm": 0.984375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.0159,
      "mean_token_accuracy": 0.09479603916406631,
      "num_tokens": 61900.0,
      "step": 35
    },
    {
      "entropy": 7.500431871414184,
      "epoch": 0.0031122349737405172,
      "grad_norm": 1.0234375,
      "learning_rate": 1.95e-05,
      "loss": 7.0403,
      "mean_token_accuracy": 0.0956958994269371,
      "num_tokens": 71234.0,
      "step": 40
    },
    {
      "entropy": 7.569026231765747,
      "epoch": 0.003501264345458082,
      "grad_norm": 1.140625,
      "learning_rate": 2.2e-05,
      "loss": 7.1142,
      "mean_token_accuracy": 0.09231705740094184,
      "num_tokens": 79610.0,
      "step": 45
    },
    {
      "entropy": 7.532996749877929,
      "epoch": 0.003890293717175647,
      "grad_norm": 1.09375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.9294,
      "mean_token_accuracy": 0.1006363645195961,
      "num_tokens": 87527.0,
      "step": 50
    },
    {
      "entropy": 7.560778427124023,
      "epoch": 0.004279323088893211,
      "grad_norm": 1.125,
      "learning_rate": 2.7e-05,
      "loss": 7.0757,
      "mean_token_accuracy": 0.09589893892407417,
      "num_tokens": 95512.0,
      "step": 55
    },
    {
      "entropy": 7.501343441009522,
      "epoch": 0.004668352460610776,
      "grad_norm": 0.99609375,
      "learning_rate": 2.95e-05,
      "loss": 7.0864,
      "mean_token_accuracy": 0.09744687899947166,
      "num_tokens": 104395.0,
      "step": 60
    },
    {
      "entropy": 7.478601932525635,
      "epoch": 0.0050573818323283405,
      "grad_norm": 0.9609375,
      "learning_rate": 3.2e-05,
      "loss": 6.9791,
      "mean_token_accuracy": 0.10184086710214615,
      "num_tokens": 112562.0,
      "step": 65
    },
    {
      "entropy": 7.405571556091308,
      "epoch": 0.005446411204045905,
      "grad_norm": 1.1015625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.86,
      "mean_token_accuracy": 0.09663667231798172,
      "num_tokens": 121962.0,
      "step": 70
    },
    {
      "entropy": 7.51507978439331,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.0234375,
      "learning_rate": 3.7e-05,
      "loss": 7.126,
      "mean_token_accuracy": 0.09231036975979805,
      "num_tokens": 130714.0,
      "step": 75
    },
    {
      "entropy": 7.520365238189697,
      "epoch": 0.0062244699474810344,
      "grad_norm": 0.95703125,
      "learning_rate": 3.95e-05,
      "loss": 6.9241,
      "mean_token_accuracy": 0.10009283497929573,
      "num_tokens": 139674.0,
      "step": 80
    },
    {
      "entropy": 7.47720103263855,
      "epoch": 0.006613499319198599,
      "grad_norm": 1.03125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.9803,
      "mean_token_accuracy": 0.09953748062252998,
      "num_tokens": 148654.0,
      "step": 85
    },
    {
      "entropy": 7.46751070022583,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.09375,
      "learning_rate": 4.45e-05,
      "loss": 6.8751,
      "mean_token_accuracy": 0.10261099562048911,
      "num_tokens": 156180.0,
      "step": 90
    },
    {
      "entropy": 7.383714389801026,
      "epoch": 0.007391558062633729,
      "grad_norm": 0.94140625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.9565,
      "mean_token_accuracy": 0.09331886917352676,
      "num_tokens": 165608.0,
      "step": 95
    },
    {
      "entropy": 7.432891321182251,
      "epoch": 0.007780587434351294,
      "grad_norm": 0.9921875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.936,
      "mean_token_accuracy": 0.10059967562556267,
      "num_tokens": 173871.0,
      "step": 100
    },
    {
      "entropy": 7.461685752868652,
      "epoch": 0.008169616806068859,
      "grad_norm": 1.0390625,
      "learning_rate": 5.2e-05,
      "loss": 7.0248,
      "mean_token_accuracy": 0.09571366757154465,
      "num_tokens": 182356.0,
      "step": 105
    },
    {
      "entropy": 7.499224996566772,
      "epoch": 0.008558646177786422,
      "grad_norm": 0.98828125,
      "learning_rate": 5.45e-05,
      "loss": 7.0179,
      "mean_token_accuracy": 0.09466839730739593,
      "num_tokens": 190361.0,
      "step": 110
    },
    {
      "entropy": 7.435716581344605,
      "epoch": 0.008947675549503988,
      "grad_norm": 1.0234375,
      "learning_rate": 5.7e-05,
      "loss": 6.9827,
      "mean_token_accuracy": 0.0984235167503357,
      "num_tokens": 199253.0,
      "step": 115
    },
    {
      "entropy": 7.429847049713135,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.109375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.9557,
      "mean_token_accuracy": 0.09870526567101479,
      "num_tokens": 208181.0,
      "step": 120
    },
    {
      "entropy": 7.515597820281982,
      "epoch": 0.009725734292939117,
      "grad_norm": 0.9921875,
      "learning_rate": 6.2e-05,
      "loss": 6.9546,
      "mean_token_accuracy": 0.09591879844665527,
      "num_tokens": 216898.0,
      "step": 125
    },
    {
      "entropy": 7.421197032928466,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.0625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.0039,
      "mean_token_accuracy": 0.09688223078846932,
      "num_tokens": 226051.0,
      "step": 130
    },
    {
      "entropy": 7.492958879470825,
      "epoch": 0.010503793036374246,
      "grad_norm": 1.015625,
      "learning_rate": 6.7e-05,
      "loss": 7.042,
      "mean_token_accuracy": 0.09841900020837784,
      "num_tokens": 235018.0,
      "step": 135
    },
    {
      "entropy": 7.432940196990967,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.0546875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.9703,
      "mean_token_accuracy": 0.09938597083091735,
      "num_tokens": 244268.0,
      "step": 140
    },
    {
      "entropy": 7.446199369430542,
      "epoch": 0.011281851779809376,
      "grad_norm": 0.98046875,
      "learning_rate": 7.2e-05,
      "loss": 6.9842,
      "mean_token_accuracy": 0.09659678265452384,
      "num_tokens": 253545.0,
      "step": 145
    },
    {
      "entropy": 7.365108060836792,
      "epoch": 0.01167088115152694,
      "grad_norm": 1.1015625,
      "learning_rate": 7.45e-05,
      "loss": 6.8048,
      "mean_token_accuracy": 0.10624724552035332,
      "num_tokens": 261767.0,
      "step": 150
    },
    {
      "entropy": 7.435336446762085,
      "epoch": 0.012059910523244505,
      "grad_norm": 0.9609375,
      "learning_rate": 7.7e-05,
      "loss": 6.9655,
      "mean_token_accuracy": 0.10120068341493607,
      "num_tokens": 270372.0,
      "step": 155
    },
    {
      "entropy": 7.463044929504394,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.046875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.026,
      "mean_token_accuracy": 0.09834155589342117,
      "num_tokens": 279726.0,
      "step": 160
    },
    {
      "entropy": 7.4832648754119875,
      "epoch": 0.012837969266679634,
      "grad_norm": 1.109375,
      "learning_rate": 8.2e-05,
      "loss": 6.9905,
      "mean_token_accuracy": 0.10089058354496956,
      "num_tokens": 288955.0,
      "step": 165
    },
    {
      "entropy": 7.449814510345459,
      "epoch": 0.013226998638397198,
      "grad_norm": 1.1171875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.9218,
      "mean_token_accuracy": 0.09420814290642739,
      "num_tokens": 297898.0,
      "step": 170
    },
    {
      "entropy": 7.396070957183838,
      "epoch": 0.013616028010114764,
      "grad_norm": 1.046875,
      "learning_rate": 8.7e-05,
      "loss": 6.9188,
      "mean_token_accuracy": 0.10343953669071197,
      "num_tokens": 306189.0,
      "step": 175
    },
    {
      "entropy": 7.427816915512085,
      "epoch": 0.014005057381832327,
      "grad_norm": 1.0703125,
      "learning_rate": 8.95e-05,
      "loss": 6.9462,
      "mean_token_accuracy": 0.09905305579304695,
      "num_tokens": 314881.0,
      "step": 180
    },
    {
      "entropy": 7.363733434677124,
      "epoch": 0.014394086753549893,
      "grad_norm": 1.1796875,
      "learning_rate": 9.2e-05,
      "loss": 6.9415,
      "mean_token_accuracy": 0.09854480773210525,
      "num_tokens": 323357.0,
      "step": 185
    },
    {
      "entropy": 7.500346422195435,
      "epoch": 0.014783116125267459,
      "grad_norm": 1.046875,
      "learning_rate": 9.45e-05,
      "loss": 6.9926,
      "mean_token_accuracy": 0.09660296440124512,
      "num_tokens": 331799.0,
      "step": 190
    },
    {
      "entropy": 7.395345211029053,
      "epoch": 0.015172145496985022,
      "grad_norm": 1.125,
      "learning_rate": 9.7e-05,
      "loss": 6.904,
      "mean_token_accuracy": 0.10265823528170585,
      "num_tokens": 340320.0,
      "step": 195
    },
    {
      "entropy": 7.404849243164063,
      "epoch": 0.015561174868702588,
      "grad_norm": 1.171875,
      "learning_rate": 9.95e-05,
      "loss": 6.8857,
      "mean_token_accuracy": 0.1047187753021717,
      "num_tokens": 348990.0,
      "step": 200
    },
    {
      "entropy": 7.39699239730835,
      "epoch": 0.01595020424042015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000102,
      "loss": 6.9065,
      "mean_token_accuracy": 0.09710254445672035,
      "num_tokens": 357661.0,
      "step": 205
    },
    {
      "entropy": 7.321025276184082,
      "epoch": 0.016339233612137717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.8787,
      "mean_token_accuracy": 0.10586154833436012,
      "num_tokens": 366099.0,
      "step": 210
    },
    {
      "entropy": 7.5005217552185055,
      "epoch": 0.016728262983855283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000107,
      "loss": 7.0379,
      "mean_token_accuracy": 0.09788483828306198,
      "num_tokens": 374926.0,
      "step": 215
    },
    {
      "entropy": 7.418247509002685,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001095,
      "loss": 6.946,
      "mean_token_accuracy": 0.098393777012825,
      "num_tokens": 383846.0,
      "step": 220
    },
    {
      "entropy": 7.378245735168457,
      "epoch": 0.01750632172729041,
      "grad_norm": 1.171875,
      "learning_rate": 0.000112,
      "loss": 6.8894,
      "mean_token_accuracy": 0.1021482452750206,
      "num_tokens": 393678.0,
      "step": 225
    },
    {
      "entropy": 7.4670178413391115,
      "epoch": 0.017895351099007976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001145,
      "loss": 6.9755,
      "mean_token_accuracy": 0.09847855567932129,
      "num_tokens": 401866.0,
      "step": 230
    },
    {
      "entropy": 7.473496532440185,
      "epoch": 0.01828438047072554,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9365,
      "mean_token_accuracy": 0.10552924722433091,
      "num_tokens": 411057.0,
      "step": 235
    },
    {
      "entropy": 7.397612524032593,
      "epoch": 0.018673409842443103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.9395,
      "mean_token_accuracy": 0.10260751843452454,
      "num_tokens": 419083.0,
      "step": 240
    },
    {
      "entropy": 7.325595569610596,
      "epoch": 0.01906243921416067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000122,
      "loss": 6.7993,
      "mean_token_accuracy": 0.10741342976689339,
      "num_tokens": 427503.0,
      "step": 245
    },
    {
      "entropy": 7.431865072250366,
      "epoch": 0.019451468585878234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001245,
      "loss": 6.9643,
      "mean_token_accuracy": 0.09718445464968681,
      "num_tokens": 435820.0,
      "step": 250
    },
    {
      "entropy": 7.301518154144287,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000127,
      "loss": 6.8424,
      "mean_token_accuracy": 0.10523844137787819,
      "num_tokens": 444692.0,
      "step": 255
    },
    {
      "entropy": 7.322385978698731,
      "epoch": 0.020229527329313362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001295,
      "loss": 6.8813,
      "mean_token_accuracy": 0.09604761153459548,
      "num_tokens": 453536.0,
      "step": 260
    },
    {
      "entropy": 7.500930261611939,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.265625,
      "learning_rate": 0.000132,
      "loss": 6.9135,
      "mean_token_accuracy": 0.10178809389472007,
      "num_tokens": 462264.0,
      "step": 265
    },
    {
      "entropy": 7.466741132736206,
      "epoch": 0.021007586072748493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.0681,
      "mean_token_accuracy": 0.09854813367128372,
      "num_tokens": 471819.0,
      "step": 270
    },
    {
      "entropy": 7.39691653251648,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.9438,
      "mean_token_accuracy": 0.0978664368391037,
      "num_tokens": 480878.0,
      "step": 275
    },
    {
      "entropy": 7.483294820785522,
      "epoch": 0.02178564481618362,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001395,
      "loss": 6.9213,
      "mean_token_accuracy": 0.1020102322101593,
      "num_tokens": 489454.0,
      "step": 280
    },
    {
      "entropy": 7.298464393615722,
      "epoch": 0.022174674187901186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8179,
      "mean_token_accuracy": 0.10413895472884178,
      "num_tokens": 498378.0,
      "step": 285
    },
    {
      "entropy": 7.501266813278198,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001445,
      "loss": 7.0766,
      "mean_token_accuracy": 0.09656166881322861,
      "num_tokens": 507638.0,
      "step": 290
    },
    {
      "entropy": 7.435027980804444,
      "epoch": 0.022952732931336317,
      "grad_norm": 1.265625,
      "learning_rate": 0.000147,
      "loss": 6.9631,
      "mean_token_accuracy": 0.10347767695784568,
      "num_tokens": 517047.0,
      "step": 295
    },
    {
      "entropy": 7.468677520751953,
      "epoch": 0.02334176230305388,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001495,
      "loss": 6.8928,
      "mean_token_accuracy": 0.10784424841403961,
      "num_tokens": 525287.0,
      "step": 300
    },
    {
      "entropy": 7.284840393066406,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000152,
      "loss": 6.827,
      "mean_token_accuracy": 0.09975214079022407,
      "num_tokens": 533266.0,
      "step": 305
    },
    {
      "entropy": 7.383808851242065,
      "epoch": 0.02411982104648901,
      "grad_norm": 1.125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.9865,
      "mean_token_accuracy": 0.09793580844998359,
      "num_tokens": 542448.0,
      "step": 310
    },
    {
      "entropy": 7.3777937412261965,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.40625,
      "learning_rate": 0.000157,
      "loss": 6.8559,
      "mean_token_accuracy": 0.10171961411833763,
      "num_tokens": 550872.0,
      "step": 315
    },
    {
      "entropy": 7.398427724838257,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001595,
      "loss": 6.945,
      "mean_token_accuracy": 0.10726686492562294,
      "num_tokens": 559616.0,
      "step": 320
    },
    {
      "entropy": 7.43606333732605,
      "epoch": 0.025286909161641703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000162,
      "loss": 6.892,
      "mean_token_accuracy": 0.1080216646194458,
      "num_tokens": 568379.0,
      "step": 325
    },
    {
      "entropy": 7.339574384689331,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.9344,
      "mean_token_accuracy": 0.09556459337472915,
      "num_tokens": 577288.0,
      "step": 330
    },
    {
      "entropy": 7.394017219543457,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8893,
      "mean_token_accuracy": 0.09673169776797294,
      "num_tokens": 587054.0,
      "step": 335
    },
    {
      "entropy": 7.3556236743927,
      "epoch": 0.026453997276794396,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0419,
      "mean_token_accuracy": 0.10046090409159661,
      "num_tokens": 595405.0,
      "step": 340
    },
    {
      "entropy": 7.431601905822754,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8507,
      "mean_token_accuracy": 0.10375126153230667,
      "num_tokens": 604095.0,
      "step": 345
    },
    {
      "entropy": 7.273375415802002,
      "epoch": 0.027232056020229527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9004,
      "mean_token_accuracy": 0.10230252891778946,
      "num_tokens": 613518.0,
      "step": 350
    },
    {
      "entropy": 7.408315944671631,
      "epoch": 0.027621085391947093,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000177,
      "loss": 6.7702,
      "mean_token_accuracy": 0.1080463670194149,
      "num_tokens": 621212.0,
      "step": 355
    },
    {
      "entropy": 7.285730218887329,
      "epoch": 0.028010114763664655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001795,
      "loss": 6.8426,
      "mean_token_accuracy": 0.1072110302746296,
      "num_tokens": 629326.0,
      "step": 360
    },
    {
      "entropy": 7.37860689163208,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.203125,
      "learning_rate": 0.000182,
      "loss": 6.8696,
      "mean_token_accuracy": 0.10251942202448845,
      "num_tokens": 637034.0,
      "step": 365
    },
    {
      "entropy": 7.277339124679566,
      "epoch": 0.028788173507099786,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001845,
      "loss": 6.7723,
      "mean_token_accuracy": 0.10722191333770752,
      "num_tokens": 645924.0,
      "step": 370
    },
    {
      "entropy": 7.407964897155762,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000187,
      "loss": 7.0054,
      "mean_token_accuracy": 0.10154261440038681,
      "num_tokens": 655020.0,
      "step": 375
    },
    {
      "entropy": 7.300928163528442,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001895,
      "loss": 6.8082,
      "mean_token_accuracy": 0.10665615424513816,
      "num_tokens": 663451.0,
      "step": 380
    },
    {
      "entropy": 7.367715358734131,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.1875,
      "learning_rate": 0.000192,
      "loss": 6.9123,
      "mean_token_accuracy": 0.1059771090745926,
      "num_tokens": 671973.0,
      "step": 385
    },
    {
      "entropy": 7.3022441387176515,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001945,
      "loss": 6.797,
      "mean_token_accuracy": 0.10909434482455253,
      "num_tokens": 680583.0,
      "step": 390
    },
    {
      "entropy": 7.2918174266815186,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8136,
      "mean_token_accuracy": 0.10621452257037163,
      "num_tokens": 689000.0,
      "step": 395
    },
    {
      "entropy": 7.341504812240601,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8056,
      "mean_token_accuracy": 0.10639143735170364,
      "num_tokens": 698047.0,
      "step": 400
    },
    {
      "entropy": 7.3614107131958,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.078125,
      "learning_rate": 0.000202,
      "loss": 6.9386,
      "mean_token_accuracy": 0.10114315450191498,
      "num_tokens": 707192.0,
      "step": 405
    },
    {
      "entropy": 7.2852061748504635,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.735,
      "mean_token_accuracy": 0.11116274818778038,
      "num_tokens": 715895.0,
      "step": 410
    },
    {
      "entropy": 7.310044527053833,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000207,
      "loss": 6.847,
      "mean_token_accuracy": 0.10309791564941406,
      "num_tokens": 724891.0,
      "step": 415
    },
    {
      "entropy": 7.340569972991943,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002095,
      "loss": 6.8936,
      "mean_token_accuracy": 0.10329565331339836,
      "num_tokens": 733378.0,
      "step": 420
    },
    {
      "entropy": 7.354770565032959,
      "epoch": 0.033067496595993,
      "grad_norm": 1.265625,
      "learning_rate": 0.000212,
      "loss": 6.875,
      "mean_token_accuracy": 0.10341520011425018,
      "num_tokens": 741098.0,
      "step": 425
    },
    {
      "entropy": 7.338204622268677,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002145,
      "loss": 6.8378,
      "mean_token_accuracy": 0.10154023990035058,
      "num_tokens": 749902.0,
      "step": 430
    },
    {
      "entropy": 7.19447169303894,
      "epoch": 0.033845555339428124,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8057,
      "mean_token_accuracy": 0.10481407195329666,
      "num_tokens": 758638.0,
      "step": 435
    },
    {
      "entropy": 7.39279146194458,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002195,
      "loss": 6.9067,
      "mean_token_accuracy": 0.10383950993418693,
      "num_tokens": 768170.0,
      "step": 440
    },
    {
      "entropy": 7.251432418823242,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000222,
      "loss": 6.7917,
      "mean_token_accuracy": 0.10390266478061676,
      "num_tokens": 776745.0,
      "step": 445
    },
    {
      "entropy": 7.344013214111328,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002245,
      "loss": 6.9049,
      "mean_token_accuracy": 0.10472740978002548,
      "num_tokens": 785030.0,
      "step": 450
    },
    {
      "entropy": 7.348719453811645,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8656,
      "mean_token_accuracy": 0.10142385959625244,
      "num_tokens": 793543.0,
      "step": 455
    },
    {
      "entropy": 7.12188196182251,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.25,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7096,
      "mean_token_accuracy": 0.1019783414900303,
      "num_tokens": 802401.0,
      "step": 460
    },
    {
      "entropy": 7.367281770706176,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8284,
      "mean_token_accuracy": 0.10576049014925956,
      "num_tokens": 811625.0,
      "step": 465
    },
    {
      "entropy": 7.186238098144531,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7775,
      "mean_token_accuracy": 0.10242061167955399,
      "num_tokens": 820742.0,
      "step": 470
    },
    {
      "entropy": 7.257758092880249,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.03125,
      "learning_rate": 0.000237,
      "loss": 6.8914,
      "mean_token_accuracy": 0.10146046429872513,
      "num_tokens": 829580.0,
      "step": 475
    },
    {
      "entropy": 7.289563941955566,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002395,
      "loss": 6.8122,
      "mean_token_accuracy": 0.10589663460850715,
      "num_tokens": 838818.0,
      "step": 480
    },
    {
      "entropy": 7.339207744598388,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000242,
      "loss": 6.8286,
      "mean_token_accuracy": 0.09875347390770912,
      "num_tokens": 847804.0,
      "step": 485
    },
    {
      "entropy": 7.0710687160491945,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.25,
      "learning_rate": 0.0002445,
      "loss": 6.5909,
      "mean_token_accuracy": 0.11272379085421562,
      "num_tokens": 856395.0,
      "step": 490
    },
    {
      "entropy": 7.308010816574097,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000247,
      "loss": 6.8648,
      "mean_token_accuracy": 0.10699901208281518,
      "num_tokens": 865219.0,
      "step": 495
    },
    {
      "entropy": 7.108584594726563,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002495,
      "loss": 6.6471,
      "mean_token_accuracy": 0.1104583702981472,
      "num_tokens": 873868.0,
      "step": 500
    },
    {
      "entropy": 7.20792145729065,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000252,
      "loss": 6.7601,
      "mean_token_accuracy": 0.10320616662502288,
      "num_tokens": 882698.0,
      "step": 505
    },
    {
      "entropy": 7.190888500213623,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002545,
      "loss": 6.6947,
      "mean_token_accuracy": 0.11407066881656647,
      "num_tokens": 891280.0,
      "step": 510
    },
    {
      "entropy": 7.244538879394531,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.25,
      "learning_rate": 0.000257,
      "loss": 6.7425,
      "mean_token_accuracy": 0.09894983395934105,
      "num_tokens": 900044.0,
      "step": 515
    },
    {
      "entropy": 7.08066873550415,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002595,
      "loss": 6.7217,
      "mean_token_accuracy": 0.0992046169936657,
      "num_tokens": 909297.0,
      "step": 520
    },
    {
      "entropy": 7.1543323516845705,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.25,
      "learning_rate": 0.000262,
      "loss": 6.7075,
      "mean_token_accuracy": 0.10772032216191292,
      "num_tokens": 917514.0,
      "step": 525
    },
    {
      "entropy": 7.194704627990722,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6951,
      "mean_token_accuracy": 0.1101997010409832,
      "num_tokens": 925842.0,
      "step": 530
    },
    {
      "entropy": 7.067590141296387,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8094,
      "mean_token_accuracy": 0.09961830377578736,
      "num_tokens": 934676.0,
      "step": 535
    },
    {
      "entropy": 7.194618606567383,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7182,
      "mean_token_accuracy": 0.10596418976783753,
      "num_tokens": 943763.0,
      "step": 540
    },
    {
      "entropy": 7.158230352401733,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.25,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7183,
      "mean_token_accuracy": 0.1088571310043335,
      "num_tokens": 952721.0,
      "step": 545
    },
    {
      "entropy": 7.143984985351563,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002745,
      "loss": 6.7143,
      "mean_token_accuracy": 0.11388995125889778,
      "num_tokens": 960899.0,
      "step": 550
    },
    {
      "entropy": 7.069172191619873,
      "epoch": 0.04318226026064968,
      "grad_norm": 1.375,
      "learning_rate": 0.000277,
      "loss": 6.7067,
      "mean_token_accuracy": 0.1085039384663105,
      "num_tokens": 969362.0,
      "step": 555
    },
    {
      "entropy": 7.178910207748413,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002795,
      "loss": 6.6775,
      "mean_token_accuracy": 0.11293213590979576,
      "num_tokens": 978028.0,
      "step": 560
    },
    {
      "entropy": 7.113482236862183,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6594,
      "mean_token_accuracy": 0.11223076805472373,
      "num_tokens": 985709.0,
      "step": 565
    },
    {
      "entropy": 7.220449066162109,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002845,
      "loss": 6.7666,
      "mean_token_accuracy": 0.10906944498419761,
      "num_tokens": 994769.0,
      "step": 570
    },
    {
      "entropy": 7.125522947311401,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.234375,
      "learning_rate": 0.000287,
      "loss": 6.7365,
      "mean_token_accuracy": 0.11124032214283944,
      "num_tokens": 1003997.0,
      "step": 575
    },
    {
      "entropy": 7.162062454223633,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002895,
      "loss": 6.7353,
      "mean_token_accuracy": 0.10934066548943519,
      "num_tokens": 1012931.0,
      "step": 580
    },
    {
      "entropy": 7.170513248443603,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000292,
      "loss": 6.7534,
      "mean_token_accuracy": 0.10608652755618095,
      "num_tokens": 1021149.0,
      "step": 585
    },
    {
      "entropy": 7.184502792358399,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002945,
      "loss": 6.7344,
      "mean_token_accuracy": 0.11254923641681672,
      "num_tokens": 1029421.0,
      "step": 590
    },
    {
      "entropy": 7.233080053329468,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.453125,
      "learning_rate": 0.000297,
      "loss": 6.8544,
      "mean_token_accuracy": 0.10738511979579926,
      "num_tokens": 1037222.0,
      "step": 595
    },
    {
      "entropy": 7.11895523071289,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002995,
      "loss": 6.7686,
      "mean_token_accuracy": 0.10719419568777085,
      "num_tokens": 1046903.0,
      "step": 600
    },
    {
      "entropy": 7.1854307651519775,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.1875,
      "learning_rate": 0.000302,
      "loss": 6.8112,
      "mean_token_accuracy": 0.10291046276688576,
      "num_tokens": 1055516.0,
      "step": 605
    },
    {
      "entropy": 7.158183145523071,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003045,
      "loss": 6.738,
      "mean_token_accuracy": 0.10869069769978523,
      "num_tokens": 1063846.0,
      "step": 610
    },
    {
      "entropy": 7.114259338378906,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000307,
      "loss": 6.7027,
      "mean_token_accuracy": 0.10939538329839707,
      "num_tokens": 1072500.0,
      "step": 615
    },
    {
      "entropy": 7.126255798339844,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003095,
      "loss": 6.7245,
      "mean_token_accuracy": 0.10950846374034881,
      "num_tokens": 1081399.0,
      "step": 620
    },
    {
      "entropy": 7.04192967414856,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000312,
      "loss": 6.6774,
      "mean_token_accuracy": 0.11167947575449944,
      "num_tokens": 1089666.0,
      "step": 625
    },
    {
      "entropy": 7.143467807769776,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003145,
      "loss": 6.5626,
      "mean_token_accuracy": 0.11535401195287705,
      "num_tokens": 1097496.0,
      "step": 630
    },
    {
      "entropy": 7.058351707458496,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000317,
      "loss": 6.7028,
      "mean_token_accuracy": 0.11161455810070038,
      "num_tokens": 1105840.0,
      "step": 635
    },
    {
      "entropy": 7.135726547241211,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003195,
      "loss": 6.7684,
      "mean_token_accuracy": 0.10728443264961243,
      "num_tokens": 1115507.0,
      "step": 640
    },
    {
      "entropy": 7.041813230514526,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.078125,
      "learning_rate": 0.000322,
      "loss": 6.7167,
      "mean_token_accuracy": 0.10211084112524986,
      "num_tokens": 1124820.0,
      "step": 645
    },
    {
      "entropy": 7.087520885467529,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5609,
      "mean_token_accuracy": 0.11316258385777474,
      "num_tokens": 1133641.0,
      "step": 650
    },
    {
      "entropy": 6.982738447189331,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.616,
      "mean_token_accuracy": 0.11105628833174705,
      "num_tokens": 1142182.0,
      "step": 655
    },
    {
      "entropy": 7.091386175155639,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6894,
      "mean_token_accuracy": 0.11387185752391815,
      "num_tokens": 1150737.0,
      "step": 660
    },
    {
      "entropy": 6.989086723327636,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5525,
      "mean_token_accuracy": 0.11128119602799416,
      "num_tokens": 1158899.0,
      "step": 665
    },
    {
      "entropy": 7.124065494537353,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6359,
      "mean_token_accuracy": 0.11705312058329582,
      "num_tokens": 1167397.0,
      "step": 670
    },
    {
      "entropy": 7.050602674484253,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.1875,
      "learning_rate": 0.000337,
      "loss": 6.6545,
      "mean_token_accuracy": 0.10680999159812928,
      "num_tokens": 1177044.0,
      "step": 675
    },
    {
      "entropy": 6.930141878128052,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003395,
      "loss": 6.524,
      "mean_token_accuracy": 0.11300303936004638,
      "num_tokens": 1185740.0,
      "step": 680
    },
    {
      "entropy": 6.976816177368164,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000342,
      "loss": 6.6738,
      "mean_token_accuracy": 0.11296947225928307,
      "num_tokens": 1194970.0,
      "step": 685
    },
    {
      "entropy": 7.070580577850341,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6134,
      "mean_token_accuracy": 0.11393867209553718,
      "num_tokens": 1203565.0,
      "step": 690
    },
    {
      "entropy": 7.023531770706176,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347,
      "loss": 6.6592,
      "mean_token_accuracy": 0.10934962257742882,
      "num_tokens": 1213226.0,
      "step": 695
    },
    {
      "entropy": 6.905843925476074,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003495,
      "loss": 6.5837,
      "mean_token_accuracy": 0.10722511857748032,
      "num_tokens": 1221710.0,
      "step": 700
    },
    {
      "entropy": 7.033835506439209,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.046875,
      "learning_rate": 0.000352,
      "loss": 6.6413,
      "mean_token_accuracy": 0.10701307356357574,
      "num_tokens": 1231125.0,
      "step": 705
    },
    {
      "entropy": 7.04109468460083,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003545,
      "loss": 6.5952,
      "mean_token_accuracy": 0.11328281164169311,
      "num_tokens": 1239521.0,
      "step": 710
    },
    {
      "entropy": 6.99030909538269,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.34375,
      "learning_rate": 0.000357,
      "loss": 6.4967,
      "mean_token_accuracy": 0.11905190572142602,
      "num_tokens": 1247363.0,
      "step": 715
    },
    {
      "entropy": 6.949197673797608,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.25,
      "learning_rate": 0.0003595,
      "loss": 6.5756,
      "mean_token_accuracy": 0.10785499513149262,
      "num_tokens": 1255395.0,
      "step": 720
    },
    {
      "entropy": 7.050080490112305,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000362,
      "loss": 6.6461,
      "mean_token_accuracy": 0.10883620008826256,
      "num_tokens": 1264135.0,
      "step": 725
    },
    {
      "entropy": 7.0190126419067385,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003645,
      "loss": 6.5703,
      "mean_token_accuracy": 0.1117468774318695,
      "num_tokens": 1272375.0,
      "step": 730
    },
    {
      "entropy": 6.988955736160278,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.125,
      "learning_rate": 0.000367,
      "loss": 6.6249,
      "mean_token_accuracy": 0.10667285323143005,
      "num_tokens": 1280960.0,
      "step": 735
    },
    {
      "entropy": 7.041416883468628,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003695,
      "loss": 6.6797,
      "mean_token_accuracy": 0.11113242357969284,
      "num_tokens": 1289273.0,
      "step": 740
    },
    {
      "entropy": 6.986820077896118,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000372,
      "loss": 6.5673,
      "mean_token_accuracy": 0.11231251433491707,
      "num_tokens": 1297571.0,
      "step": 745
    },
    {
      "entropy": 7.002435827255249,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003745,
      "loss": 6.6165,
      "mean_token_accuracy": 0.11419098302721978,
      "num_tokens": 1305686.0,
      "step": 750
    },
    {
      "entropy": 6.96313853263855,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.1875,
      "learning_rate": 0.000377,
      "loss": 6.6372,
      "mean_token_accuracy": 0.1087796851992607,
      "num_tokens": 1314483.0,
      "step": 755
    },
    {
      "entropy": 6.946442985534668,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003795,
      "loss": 6.5161,
      "mean_token_accuracy": 0.10849898308515549,
      "num_tokens": 1323183.0,
      "step": 760
    },
    {
      "entropy": 6.952772951126098,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000382,
      "loss": 6.6204,
      "mean_token_accuracy": 0.11015321984887123,
      "num_tokens": 1331964.0,
      "step": 765
    },
    {
      "entropy": 6.933269834518432,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003845,
      "loss": 6.5707,
      "mean_token_accuracy": 0.11390582174062729,
      "num_tokens": 1340185.0,
      "step": 770
    },
    {
      "entropy": 7.053455209732055,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.6715,
      "mean_token_accuracy": 0.1104957066476345,
      "num_tokens": 1349522.0,
      "step": 775
    },
    {
      "entropy": 7.057522010803223,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6406,
      "mean_token_accuracy": 0.10729108974337578,
      "num_tokens": 1357893.0,
      "step": 780
    },
    {
      "entropy": 6.856075191497803,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5272,
      "mean_token_accuracy": 0.11494160592555999,
      "num_tokens": 1366685.0,
      "step": 785
    },
    {
      "entropy": 6.961356449127197,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4661,
      "mean_token_accuracy": 0.11778462678194046,
      "num_tokens": 1375245.0,
      "step": 790
    },
    {
      "entropy": 6.9094297885894775,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.6459,
      "mean_token_accuracy": 0.1119820773601532,
      "num_tokens": 1383724.0,
      "step": 795
    },
    {
      "entropy": 7.064331769943237,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003995,
      "loss": 6.6246,
      "mean_token_accuracy": 0.11086539849638939,
      "num_tokens": 1392455.0,
      "step": 800
    },
    {
      "entropy": 7.040333318710327,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000402,
      "loss": 6.6322,
      "mean_token_accuracy": 0.11222699806094169,
      "num_tokens": 1401262.0,
      "step": 805
    },
    {
      "entropy": 6.833669853210449,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004045,
      "loss": 6.528,
      "mean_token_accuracy": 0.11181755363941193,
      "num_tokens": 1409991.0,
      "step": 810
    },
    {
      "entropy": 6.826368951797486,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5143,
      "mean_token_accuracy": 0.11217057034373283,
      "num_tokens": 1418208.0,
      "step": 815
    },
    {
      "entropy": 7.009344673156738,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004095,
      "loss": 6.5279,
      "mean_token_accuracy": 0.1156628392636776,
      "num_tokens": 1426665.0,
      "step": 820
    },
    {
      "entropy": 6.799173259735108,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000412,
      "loss": 6.4582,
      "mean_token_accuracy": 0.11917970553040505,
      "num_tokens": 1435247.0,
      "step": 825
    },
    {
      "entropy": 6.870002889633179,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004145,
      "loss": 6.5052,
      "mean_token_accuracy": 0.11340074241161346,
      "num_tokens": 1443678.0,
      "step": 830
    },
    {
      "entropy": 6.767337274551392,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000417,
      "loss": 6.4409,
      "mean_token_accuracy": 0.12152031734585762,
      "num_tokens": 1451567.0,
      "step": 835
    },
    {
      "entropy": 6.837034368515015,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004195,
      "loss": 6.4578,
      "mean_token_accuracy": 0.11555799394845963,
      "num_tokens": 1460618.0,
      "step": 840
    },
    {
      "entropy": 6.84786286354065,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.203125,
      "learning_rate": 0.000422,
      "loss": 6.521,
      "mean_token_accuracy": 0.1165824718773365,
      "num_tokens": 1469821.0,
      "step": 845
    },
    {
      "entropy": 6.948866891860962,
      "epoch": 0.066134993191986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004245,
      "loss": 6.596,
      "mean_token_accuracy": 0.11146286725997925,
      "num_tokens": 1477989.0,
      "step": 850
    },
    {
      "entropy": 6.937360382080078,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.203125,
      "learning_rate": 0.000427,
      "loss": 6.6151,
      "mean_token_accuracy": 0.11400183513760567,
      "num_tokens": 1486822.0,
      "step": 855
    },
    {
      "entropy": 6.887311553955078,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004295,
      "loss": 6.4833,
      "mean_token_accuracy": 0.11117951571941376,
      "num_tokens": 1496604.0,
      "step": 860
    },
    {
      "entropy": 6.801223611831665,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000432,
      "loss": 6.5104,
      "mean_token_accuracy": 0.11625081971287728,
      "num_tokens": 1504725.0,
      "step": 865
    },
    {
      "entropy": 6.889442777633667,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004345,
      "loss": 6.5156,
      "mean_token_accuracy": 0.11729802563786507,
      "num_tokens": 1513222.0,
      "step": 870
    },
    {
      "entropy": 6.880262184143066,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000437,
      "loss": 6.4837,
      "mean_token_accuracy": 0.1164153091609478,
      "num_tokens": 1522183.0,
      "step": 875
    },
    {
      "entropy": 6.819140386581421,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004395,
      "loss": 6.5135,
      "mean_token_accuracy": 0.11310733184218406,
      "num_tokens": 1531316.0,
      "step": 880
    },
    {
      "entropy": 6.924764108657837,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.265625,
      "learning_rate": 0.000442,
      "loss": 6.579,
      "mean_token_accuracy": 0.11591333150863647,
      "num_tokens": 1540053.0,
      "step": 885
    },
    {
      "entropy": 6.83437933921814,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004445,
      "loss": 6.4385,
      "mean_token_accuracy": 0.1197456382215023,
      "num_tokens": 1549391.0,
      "step": 890
    },
    {
      "entropy": 6.9253668785095215,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.25,
      "learning_rate": 0.000447,
      "loss": 6.5571,
      "mean_token_accuracy": 0.1145814374089241,
      "num_tokens": 1557364.0,
      "step": 895
    },
    {
      "entropy": 6.799445247650146,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3947,
      "mean_token_accuracy": 0.1110591433942318,
      "num_tokens": 1567011.0,
      "step": 900
    },
    {
      "entropy": 6.749784517288208,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5116,
      "mean_token_accuracy": 0.11603075638413429,
      "num_tokens": 1575441.0,
      "step": 905
    },
    {
      "entropy": 6.821898317337036,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4575,
      "mean_token_accuracy": 0.11487730145454407,
      "num_tokens": 1583757.0,
      "step": 910
    },
    {
      "entropy": 6.860511064529419,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5019,
      "mean_token_accuracy": 0.11874779388308525,
      "num_tokens": 1592406.0,
      "step": 915
    },
    {
      "entropy": 6.792093992233276,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4787,
      "mean_token_accuracy": 0.11637907922267914,
      "num_tokens": 1601665.0,
      "step": 920
    },
    {
      "entropy": 6.8411633491516115,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.46875,
      "learning_rate": 0.000462,
      "loss": 6.502,
      "mean_token_accuracy": 0.11683260723948478,
      "num_tokens": 1609788.0,
      "step": 925
    },
    {
      "entropy": 6.7901835441589355,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004645,
      "loss": 6.5025,
      "mean_token_accuracy": 0.11982274875044822,
      "num_tokens": 1618030.0,
      "step": 930
    },
    {
      "entropy": 6.767443609237671,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000467,
      "loss": 6.5022,
      "mean_token_accuracy": 0.11648002341389656,
      "num_tokens": 1626901.0,
      "step": 935
    },
    {
      "entropy": 6.773926687240601,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004695,
      "loss": 6.3908,
      "mean_token_accuracy": 0.1169219508767128,
      "num_tokens": 1635372.0,
      "step": 940
    },
    {
      "entropy": 6.745931243896484,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472,
      "loss": 6.4347,
      "mean_token_accuracy": 0.1165583461523056,
      "num_tokens": 1643791.0,
      "step": 945
    },
    {
      "entropy": 6.671934700012207,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.25,
      "learning_rate": 0.0004745,
      "loss": 6.4064,
      "mean_token_accuracy": 0.12068933174014092,
      "num_tokens": 1652161.0,
      "step": 950
    },
    {
      "entropy": 6.832337713241577,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000477,
      "loss": 6.4351,
      "mean_token_accuracy": 0.11309069246053696,
      "num_tokens": 1661077.0,
      "step": 955
    },
    {
      "entropy": 6.688000345230103,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004795,
      "loss": 6.4195,
      "mean_token_accuracy": 0.11566439867019654,
      "num_tokens": 1669010.0,
      "step": 960
    },
    {
      "entropy": 6.826863241195679,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.15625,
      "learning_rate": 0.000482,
      "loss": 6.5798,
      "mean_token_accuracy": 0.11118761152029037,
      "num_tokens": 1678002.0,
      "step": 965
    },
    {
      "entropy": 6.840054130554199,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004845,
      "loss": 6.532,
      "mean_token_accuracy": 0.11202086359262467,
      "num_tokens": 1687403.0,
      "step": 970
    },
    {
      "entropy": 6.791687536239624,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.203125,
      "learning_rate": 0.000487,
      "loss": 6.4514,
      "mean_token_accuracy": 0.11927441358566285,
      "num_tokens": 1695512.0,
      "step": 975
    },
    {
      "entropy": 6.763000774383545,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895,
      "loss": 6.4496,
      "mean_token_accuracy": 0.1148236095905304,
      "num_tokens": 1705116.0,
      "step": 980
    },
    {
      "entropy": 6.659470129013061,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492,
      "loss": 6.3587,
      "mean_token_accuracy": 0.11876563057303428,
      "num_tokens": 1713715.0,
      "step": 985
    },
    {
      "entropy": 6.745743465423584,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004945,
      "loss": 6.4189,
      "mean_token_accuracy": 0.11852910742163658,
      "num_tokens": 1721881.0,
      "step": 990
    },
    {
      "entropy": 6.7183277130126955,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497,
      "loss": 6.4432,
      "mean_token_accuracy": 0.1176742009818554,
      "num_tokens": 1730716.0,
      "step": 995
    },
    {
      "entropy": 6.888888120651245,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995,
      "loss": 6.5998,
      "mean_token_accuracy": 0.11513158455491065,
      "num_tokens": 1740257.0,
      "step": 1000
    },
    {
      "entropy": 6.787261962890625,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.5037,
      "mean_token_accuracy": 0.1148828312754631,
      "num_tokens": 1749071.0,
      "step": 1005
    },
    {
      "entropy": 6.743206596374511,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.5174,
      "mean_token_accuracy": 0.10950450450181962,
      "num_tokens": 1757952.0,
      "step": 1010
    },
    {
      "entropy": 6.725841903686524,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.125,
      "learning_rate": 0.000499999986617054,
      "loss": 6.4453,
      "mean_token_accuracy": 0.11559172496199607,
      "num_tokens": 1766027.0,
      "step": 1015
    },
    {
      "entropy": 6.709922885894775,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.4473,
      "mean_token_accuracy": 0.11136003509163857,
      "num_tokens": 1775039.0,
      "step": 1020
    },
    {
      "entropy": 6.766154432296753,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.4508,
      "mean_token_accuracy": 0.12302565276622772,
      "num_tokens": 1783971.0,
      "step": 1025
    },
    {
      "entropy": 6.6978371143341064,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.3723,
      "mean_token_accuracy": 0.11690968126058579,
      "num_tokens": 1792743.0,
      "step": 1030
    },
    {
      "entropy": 6.746917724609375,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.5249,
      "mean_token_accuracy": 0.10790274739265442,
      "num_tokens": 1801435.0,
      "step": 1035
    },
    {
      "entropy": 6.829019927978516,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.4185,
      "mean_token_accuracy": 0.11439956873655319,
      "num_tokens": 1810147.0,
      "step": 1040
    },
    {
      "entropy": 6.664861822128296,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.3459,
      "mean_token_accuracy": 0.12141675725579262,
      "num_tokens": 1818672.0,
      "step": 1045
    },
    {
      "entropy": 6.779337644577026,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.5038,
      "mean_token_accuracy": 0.11229550465941429,
      "num_tokens": 1827590.0,
      "step": 1050
    },
    {
      "entropy": 6.68260350227356,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499999800894565,
      "loss": 6.4138,
      "mean_token_accuracy": 0.11804961189627647,
      "num_tokens": 1836467.0,
      "step": 1055
    },
    {
      "entropy": 6.659322786331177,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.3545,
      "mean_token_accuracy": 0.11814910471439362,
      "num_tokens": 1845537.0,
      "step": 1060
    },
    {
      "entropy": 6.668975210189819,
      "epoch": 0.08286325617584128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.3591,
      "mean_token_accuracy": 0.12657058462500573,
      "num_tokens": 1854433.0,
      "step": 1065
    },
    {
      "entropy": 6.699471855163575,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499999674917391,
      "loss": 6.4629,
      "mean_token_accuracy": 0.11414268612861633,
      "num_tokens": 1863708.0,
      "step": 1070
    },
    {
      "entropy": 6.6790022373199465,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.3401,
      "mean_token_accuracy": 0.12237241193652153,
      "num_tokens": 1871854.0,
      "step": 1075
    },
    {
      "entropy": 6.618637943267823,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.2953,
      "mean_token_accuracy": 0.12108489871025085,
      "num_tokens": 1880471.0,
      "step": 1080
    },
    {
      "entropy": 6.6455847263336185,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.3643,
      "mean_token_accuracy": 0.12223557531833648,
      "num_tokens": 1889882.0,
      "step": 1085
    },
    {
      "entropy": 6.545784616470337,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.255,
      "mean_token_accuracy": 0.12821755334734916,
      "num_tokens": 1898701.0,
      "step": 1090
    },
    {
      "entropy": 6.611828470230103,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.3715,
      "mean_token_accuracy": 0.11571631282567978,
      "num_tokens": 1908165.0,
      "step": 1095
    },
    {
      "entropy": 6.640139150619507,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499999330784743,
      "loss": 6.4811,
      "mean_token_accuracy": 0.1190013438463211,
      "num_tokens": 1916295.0,
      "step": 1100
    },
    {
      "entropy": 6.72911229133606,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.3657,
      "mean_token_accuracy": 0.12518487498164177,
      "num_tokens": 1924643.0,
      "step": 1105
    },
    {
      "entropy": 6.63273024559021,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.3742,
      "mean_token_accuracy": 0.11607154682278634,
      "num_tokens": 1933822.0,
      "step": 1110
    },
    {
      "entropy": 6.567139911651611,
      "epoch": 0.08675354989301692,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.3671,
      "mean_token_accuracy": 0.119285686314106,
      "num_tokens": 1943278.0,
      "step": 1115
    },
    {
      "entropy": 6.764870882034302,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.4038,
      "mean_token_accuracy": 0.11785744428634644,
      "num_tokens": 1951883.0,
      "step": 1120
    },
    {
      "entropy": 6.537186765670777,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.2868,
      "mean_token_accuracy": 0.12786889150738717,
      "num_tokens": 1959567.0,
      "step": 1125
    },
    {
      "entropy": 6.700858688354492,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.3722,
      "mean_token_accuracy": 0.12128480300307273,
      "num_tokens": 1968190.0,
      "step": 1130
    },
    {
      "entropy": 6.507136249542237,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.2919,
      "mean_token_accuracy": 0.1241019494831562,
      "num_tokens": 1977519.0,
      "step": 1135
    },
    {
      "entropy": 6.6155976295471195,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.2728,
      "mean_token_accuracy": 0.12157176062464714,
      "num_tokens": 1985796.0,
      "step": 1140
    },
    {
      "entropy": 6.703161907196045,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.4198,
      "mean_token_accuracy": 0.11606316342949867,
      "num_tokens": 1993874.0,
      "step": 1145
    },
    {
      "entropy": 6.698982381820679,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.3686,
      "mean_token_accuracy": 0.12184037938714028,
      "num_tokens": 2001663.0,
      "step": 1150
    },
    {
      "entropy": 6.559295749664306,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499998380665456,
      "loss": 6.3539,
      "mean_token_accuracy": 0.1229154959321022,
      "num_tokens": 2010179.0,
      "step": 1155
    },
    {
      "entropy": 6.653844833374023,
      "epoch": 0.090254814238475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.2589,
      "mean_token_accuracy": 0.12252619490027428,
      "num_tokens": 2020013.0,
      "step": 1160
    },
    {
      "entropy": 6.594068193435669,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.3751,
      "mean_token_accuracy": 0.12590979486703874,
      "num_tokens": 2028396.0,
      "step": 1165
    },
    {
      "entropy": 6.726529788970947,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.4733,
      "mean_token_accuracy": 0.11403553262352943,
      "num_tokens": 2036947.0,
      "step": 1170
    },
    {
      "entropy": 6.585730981826782,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.3254,
      "mean_token_accuracy": 0.11759492009878159,
      "num_tokens": 2045162.0,
      "step": 1175
    },
    {
      "entropy": 6.6485522270202635,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.4095,
      "mean_token_accuracy": 0.11743757203221321,
      "num_tokens": 2054198.0,
      "step": 1180
    },
    {
      "entropy": 6.583405160903931,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.3297,
      "mean_token_accuracy": 0.11797131523489952,
      "num_tokens": 2062558.0,
      "step": 1185
    },
    {
      "entropy": 6.551700019836426,
      "epoch": 0.0925889904687804,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.2095,
      "mean_token_accuracy": 0.11727948859333992,
      "num_tokens": 2070816.0,
      "step": 1190
    },
    {
      "entropy": 6.552899169921875,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.2712,
      "mean_token_accuracy": 0.12385502606630325,
      "num_tokens": 2079058.0,
      "step": 1195
    },
    {
      "entropy": 6.530488681793213,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.2343,
      "mean_token_accuracy": 0.12390355318784714,
      "num_tokens": 2086679.0,
      "step": 1200
    },
    {
      "entropy": 6.622032976150512,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.3462,
      "mean_token_accuracy": 0.12203274443745613,
      "num_tokens": 2095788.0,
      "step": 1205
    },
    {
      "entropy": 6.5792169094085695,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.3628,
      "mean_token_accuracy": 0.11934652253985405,
      "num_tokens": 2105435.0,
      "step": 1210
    },
    {
      "entropy": 6.570254611968994,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.3156,
      "mean_token_accuracy": 0.1212031789124012,
      "num_tokens": 2114318.0,
      "step": 1215
    },
    {
      "entropy": 6.521538877487183,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.245,
      "mean_token_accuracy": 0.13023234903812408,
      "num_tokens": 2122707.0,
      "step": 1220
    },
    {
      "entropy": 6.497583961486816,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.3365,
      "mean_token_accuracy": 0.1240180067718029,
      "num_tokens": 2131176.0,
      "step": 1225
    },
    {
      "entropy": 6.681749296188355,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.3785,
      "mean_token_accuracy": 0.12624216973781585,
      "num_tokens": 2139872.0,
      "step": 1230
    },
    {
      "entropy": 6.508302974700928,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.2386,
      "mean_token_accuracy": 0.1306472271680832,
      "num_tokens": 2148668.0,
      "step": 1235
    },
    {
      "entropy": 6.534418439865112,
      "epoch": 0.09647928418595604,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.24,
      "mean_token_accuracy": 0.12787432298064233,
      "num_tokens": 2156628.0,
      "step": 1240
    },
    {
      "entropy": 6.563996267318726,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.3335,
      "mean_token_accuracy": 0.11790938824415206,
      "num_tokens": 2165620.0,
      "step": 1245
    },
    {
      "entropy": 6.626491832733154,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.3967,
      "mean_token_accuracy": 0.11869805678725243,
      "num_tokens": 2174493.0,
      "step": 1250
    },
    {
      "entropy": 6.55765929222107,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499995594840114,
      "loss": 6.2909,
      "mean_token_accuracy": 0.11683699488639832,
      "num_tokens": 2183258.0,
      "step": 1255
    },
    {
      "entropy": 6.548435878753662,
      "epoch": 0.0980354016728263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.3454,
      "mean_token_accuracy": 0.11895403861999512,
      "num_tokens": 2192263.0,
      "step": 1260
    },
    {
      "entropy": 6.651948308944702,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.3889,
      "mean_token_accuracy": 0.1177945502102375,
      "num_tokens": 2201797.0,
      "step": 1265
    },
    {
      "entropy": 6.354124736785889,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.1299,
      "mean_token_accuracy": 0.1311613693833351,
      "num_tokens": 2209454.0,
      "step": 1270
    },
    {
      "entropy": 6.649595832824707,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.3022,
      "mean_token_accuracy": 0.12611282840371132,
      "num_tokens": 2217761.0,
      "step": 1275
    },
    {
      "entropy": 6.47292742729187,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.2906,
      "mean_token_accuracy": 0.12207216396927834,
      "num_tokens": 2225793.0,
      "step": 1280
    },
    {
      "entropy": 6.592188453674316,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.1805,
      "mean_token_accuracy": 0.1319039449095726,
      "num_tokens": 2234620.0,
      "step": 1285
    },
    {
      "entropy": 6.4829436302185055,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.125,
      "learning_rate": 0.000499994297182014,
      "loss": 6.2643,
      "mean_token_accuracy": 0.12326281145215034,
      "num_tokens": 2242829.0,
      "step": 1290
    },
    {
      "entropy": 6.57083477973938,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.3014,
      "mean_token_accuracy": 0.11634257808327675,
      "num_tokens": 2251356.0,
      "step": 1295
    },
    {
      "entropy": 6.523905658721924,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.3684,
      "mean_token_accuracy": 0.11852610781788826,
      "num_tokens": 2259327.0,
      "step": 1300
    },
    {
      "entropy": 6.59006552696228,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.3871,
      "mean_token_accuracy": 0.11884624063968659,
      "num_tokens": 2267948.0,
      "step": 1305
    },
    {
      "entropy": 6.446884679794311,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.194,
      "mean_token_accuracy": 0.1225943498313427,
      "num_tokens": 2275828.0,
      "step": 1310
    },
    {
      "entropy": 6.5004541873931885,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499993267865406,
      "loss": 6.1373,
      "mean_token_accuracy": 0.1289021112024784,
      "num_tokens": 2284246.0,
      "step": 1315
    },
    {
      "entropy": 6.385093450546265,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.1261,
      "mean_token_accuracy": 0.13380536437034607,
      "num_tokens": 2291837.0,
      "step": 1320
    },
    {
      "entropy": 6.546138572692871,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.206,
      "mean_token_accuracy": 0.12851327955722808,
      "num_tokens": 2299853.0,
      "step": 1325
    },
    {
      "entropy": 6.574787330627442,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.3075,
      "mean_token_accuracy": 0.1206931158900261,
      "num_tokens": 2308416.0,
      "step": 1330
    },
    {
      "entropy": 6.507684230804443,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.3388,
      "mean_token_accuracy": 0.12228097319602967,
      "num_tokens": 2318104.0,
      "step": 1335
    },
    {
      "entropy": 6.501170110702515,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.2612,
      "mean_token_accuracy": 0.12383703589439392,
      "num_tokens": 2326807.0,
      "step": 1340
    },
    {
      "entropy": 6.528704977035522,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.259,
      "mean_token_accuracy": 0.12394311875104905,
      "num_tokens": 2335921.0,
      "step": 1345
    },
    {
      "entropy": 6.505633020401001,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.2739,
      "mean_token_accuracy": 0.12304315939545632,
      "num_tokens": 2344914.0,
      "step": 1350
    },
    {
      "entropy": 6.447750616073608,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.2044,
      "mean_token_accuracy": 0.125525876134634,
      "num_tokens": 2353211.0,
      "step": 1355
    },
    {
      "entropy": 6.4893646240234375,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.3487,
      "mean_token_accuracy": 0.12199961394071579,
      "num_tokens": 2361717.0,
      "step": 1360
    },
    {
      "entropy": 6.525090074539184,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499990953189016,
      "loss": 6.2712,
      "mean_token_accuracy": 0.13059643879532815,
      "num_tokens": 2370114.0,
      "step": 1365
    },
    {
      "entropy": 6.630150699615479,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.263,
      "mean_token_accuracy": 0.12237593159079552,
      "num_tokens": 2378820.0,
      "step": 1370
    },
    {
      "entropy": 6.354615974426269,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.1715,
      "mean_token_accuracy": 0.12704389467835425,
      "num_tokens": 2387119.0,
      "step": 1375
    },
    {
      "entropy": 6.53058009147644,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.2259,
      "mean_token_accuracy": 0.12630244940519333,
      "num_tokens": 2396246.0,
      "step": 1380
    },
    {
      "entropy": 6.406842041015625,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.2855,
      "mean_token_accuracy": 0.12351492792367935,
      "num_tokens": 2405428.0,
      "step": 1385
    },
    {
      "entropy": 6.490512895584106,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.2421,
      "mean_token_accuracy": 0.1267989382147789,
      "num_tokens": 2414669.0,
      "step": 1390
    },
    {
      "entropy": 6.563437604904175,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.3,
      "mean_token_accuracy": 0.12090388685464859,
      "num_tokens": 2423193.0,
      "step": 1395
    },
    {
      "entropy": 6.500105094909668,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499989129789502,
      "loss": 6.2664,
      "mean_token_accuracy": 0.12048013284802436,
      "num_tokens": 2431150.0,
      "step": 1400
    },
    {
      "entropy": 6.445392656326294,
      "epoch": 0.10931725345263567,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.2446,
      "mean_token_accuracy": 0.12127585709095001,
      "num_tokens": 2440337.0,
      "step": 1405
    },
    {
      "entropy": 6.441078138351441,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.1736,
      "mean_token_accuracy": 0.1296399176120758,
      "num_tokens": 2449246.0,
      "step": 1410
    },
    {
      "entropy": 6.434943819046021,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.1752,
      "mean_token_accuracy": 0.13090624362230302,
      "num_tokens": 2457982.0,
      "step": 1415
    },
    {
      "entropy": 6.44070029258728,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.1661,
      "mean_token_accuracy": 0.12823255807161332,
      "num_tokens": 2466697.0,
      "step": 1420
    },
    {
      "entropy": 6.419079399108886,
      "epoch": 0.11087337093950593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.1397,
      "mean_token_accuracy": 0.13144427835941314,
      "num_tokens": 2475457.0,
      "step": 1425
    },
    {
      "entropy": 6.336713647842407,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.1393,
      "mean_token_accuracy": 0.12938348650932313,
      "num_tokens": 2484753.0,
      "step": 1430
    },
    {
      "entropy": 6.465570974349975,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.2556,
      "mean_token_accuracy": 0.124443519115448,
      "num_tokens": 2492921.0,
      "step": 1435
    },
    {
      "entropy": 6.46670618057251,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.224,
      "mean_token_accuracy": 0.12451541647315026,
      "num_tokens": 2501371.0,
      "step": 1440
    },
    {
      "entropy": 6.471292209625244,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499986539621578,
      "loss": 6.1023,
      "mean_token_accuracy": 0.12834953069686889,
      "num_tokens": 2509521.0,
      "step": 1445
    },
    {
      "entropy": 6.416775035858154,
      "epoch": 0.11281851779809375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.2143,
      "mean_token_accuracy": 0.12635174617171288,
      "num_tokens": 2518316.0,
      "step": 1450
    },
    {
      "entropy": 6.514144420623779,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.309,
      "mean_token_accuracy": 0.1220340982079506,
      "num_tokens": 2527549.0,
      "step": 1455
    },
    {
      "entropy": 6.501974773406983,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.33,
      "mean_token_accuracy": 0.12365049570798874,
      "num_tokens": 2536341.0,
      "step": 1460
    },
    {
      "entropy": 6.390389823913575,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.2144,
      "mean_token_accuracy": 0.13640521019697188,
      "num_tokens": 2544869.0,
      "step": 1465
    },
    {
      "entropy": 6.442426967620849,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.1124,
      "mean_token_accuracy": 0.1380823515355587,
      "num_tokens": 2553432.0,
      "step": 1470
    },
    {
      "entropy": 6.435215473175049,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.245,
      "mean_token_accuracy": 0.12433946654200553,
      "num_tokens": 2561996.0,
      "step": 1475
    },
    {
      "entropy": 6.435883617401123,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.1732,
      "mean_token_accuracy": 0.13035295605659486,
      "num_tokens": 2571315.0,
      "step": 1480
    },
    {
      "entropy": 6.433837747573852,
      "epoch": 0.1155417234001167,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.2935,
      "mean_token_accuracy": 0.12482402846217155,
      "num_tokens": 2579589.0,
      "step": 1485
    },
    {
      "entropy": 6.433605575561524,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.0607,
      "mean_token_accuracy": 0.1283522941172123,
      "num_tokens": 2588528.0,
      "step": 1490
    },
    {
      "entropy": 6.3824395656585695,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999833373452,
      "loss": 6.1744,
      "mean_token_accuracy": 0.12705983147025107,
      "num_tokens": 2596820.0,
      "step": 1495
    },
    {
      "entropy": 6.30913896560669,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.1168,
      "mean_token_accuracy": 0.12735813558101655,
      "num_tokens": 2605502.0,
      "step": 1500
    },
    {
      "entropy": 6.378449249267578,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.1193,
      "mean_token_accuracy": 0.13392857909202577,
      "num_tokens": 2613974.0,
      "step": 1505
    },
    {
      "entropy": 6.370792722702026,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.1965,
      "mean_token_accuracy": 0.12880947291851044,
      "num_tokens": 2622581.0,
      "step": 1510
    },
    {
      "entropy": 6.402956151962281,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.0841,
      "mean_token_accuracy": 0.1251236990094185,
      "num_tokens": 2631868.0,
      "step": 1515
    },
    {
      "entropy": 6.339415788650513,
      "epoch": 0.11826492900213967,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.1093,
      "mean_token_accuracy": 0.1355234570801258,
      "num_tokens": 2640501.0,
      "step": 1520
    },
    {
      "entropy": 6.433193206787109,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.1035,
      "mean_token_accuracy": 0.13361863121390344,
      "num_tokens": 2648916.0,
      "step": 1525
    },
    {
      "entropy": 6.406140279769898,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.1896,
      "mean_token_accuracy": 0.1334061346948147,
      "num_tokens": 2657725.0,
      "step": 1530
    },
    {
      "entropy": 6.353977298736572,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.1751,
      "mean_token_accuracy": 0.1281532824039459,
      "num_tokens": 2665983.0,
      "step": 1535
    },
    {
      "entropy": 6.392253303527832,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.1957,
      "mean_token_accuracy": 0.13333539366722108,
      "num_tokens": 2674546.0,
      "step": 1540
    },
    {
      "entropy": 6.389540004730224,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.1091,
      "mean_token_accuracy": 0.13411504104733468,
      "num_tokens": 2684088.0,
      "step": 1545
    },
    {
      "entropy": 6.37681622505188,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499979420551689,
      "loss": 6.1664,
      "mean_token_accuracy": 0.12699728310108185,
      "num_tokens": 2693289.0,
      "step": 1550
    },
    {
      "entropy": 6.318667316436768,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499979043997213,
      "loss": 6.095,
      "mean_token_accuracy": 0.13104904145002366,
      "num_tokens": 2701744.0,
      "step": 1555
    },
    {
      "entropy": 6.411156368255615,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.1499,
      "mean_token_accuracy": 0.12439182624220849,
      "num_tokens": 2710584.0,
      "step": 1560
    },
    {
      "entropy": 6.411997509002686,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.1491,
      "mean_token_accuracy": 0.13031079471111298,
      "num_tokens": 2719478.0,
      "step": 1565
    },
    {
      "entropy": 6.465567684173584,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.2286,
      "mean_token_accuracy": 0.12305473163723946,
      "num_tokens": 2729105.0,
      "step": 1570
    },
    {
      "entropy": 6.401231145858764,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499977503642378,
      "loss": 6.1908,
      "mean_token_accuracy": 0.1272389329969883,
      "num_tokens": 2738300.0,
      "step": 1575
    },
    {
      "entropy": 6.384681034088135,
      "epoch": 0.12293328146275044,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.1335,
      "mean_token_accuracy": 0.12494338452816009,
      "num_tokens": 2747236.0,
      "step": 1580
    },
    {
      "entropy": 6.289424848556519,
      "epoch": 0.123322310834468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.0956,
      "mean_token_accuracy": 0.13419784530997275,
      "num_tokens": 2755919.0,
      "step": 1585
    },
    {
      "entropy": 6.305175161361694,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.0768,
      "mean_token_accuracy": 0.1307978369295597,
      "num_tokens": 2764469.0,
      "step": 1590
    },
    {
      "entropy": 6.355486297607422,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.081,
      "mean_token_accuracy": 0.1366819642484188,
      "num_tokens": 2772663.0,
      "step": 1595
    },
    {
      "entropy": 6.419379472732544,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.1161,
      "mean_token_accuracy": 0.12876046895980836,
      "num_tokens": 2780719.0,
      "step": 1600
    },
    {
      "entropy": 6.330186557769776,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499975090699982,
      "loss": 6.1714,
      "mean_token_accuracy": 0.128857558965683,
      "num_tokens": 2789186.0,
      "step": 1605
    },
    {
      "entropy": 6.337528562545776,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.0856,
      "mean_token_accuracy": 0.13346567153930664,
      "num_tokens": 2797993.0,
      "step": 1610
    },
    {
      "entropy": 6.312220811843872,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.0464,
      "mean_token_accuracy": 0.13435178622603416,
      "num_tokens": 2806375.0,
      "step": 1615
    },
    {
      "entropy": 6.440674829483032,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.1942,
      "mean_token_accuracy": 0.13562749326229095,
      "num_tokens": 2814772.0,
      "step": 1620
    },
    {
      "entropy": 6.202065896987915,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.0582,
      "mean_token_accuracy": 0.13596921488642694,
      "num_tokens": 2824231.0,
      "step": 1625
    },
    {
      "entropy": 6.376500511169434,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499972986039528,
      "loss": 6.1245,
      "mean_token_accuracy": 0.12782634794712067,
      "num_tokens": 2833470.0,
      "step": 1630
    },
    {
      "entropy": 6.353403902053833,
      "epoch": 0.12721260455164365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.1327,
      "mean_token_accuracy": 0.1281338281929493,
      "num_tokens": 2843010.0,
      "step": 1635
    },
    {
      "entropy": 6.346253395080566,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.0231,
      "mean_token_accuracy": 0.13678104057908058,
      "num_tokens": 2851545.0,
      "step": 1640
    },
    {
      "entropy": 6.269683694839477,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.1796,
      "mean_token_accuracy": 0.11961811035871506,
      "num_tokens": 2861230.0,
      "step": 1645
    },
    {
      "entropy": 6.395336675643921,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.1356,
      "mean_token_accuracy": 0.1289409004151821,
      "num_tokens": 2869436.0,
      "step": 1650
    },
    {
      "entropy": 6.306952619552613,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999707960389021,
      "loss": 5.9996,
      "mean_token_accuracy": 0.13151076734066008,
      "num_tokens": 2877688.0,
      "step": 1655
    },
    {
      "entropy": 6.365733051300049,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.1389,
      "mean_token_accuracy": 0.13166750520467757,
      "num_tokens": 2887043.0,
      "step": 1660
    },
    {
      "entropy": 6.262747430801392,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.0721,
      "mean_token_accuracy": 0.13516117706894876,
      "num_tokens": 2895732.0,
      "step": 1665
    },
    {
      "entropy": 6.256302070617676,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.039,
      "mean_token_accuracy": 0.13851652070879936,
      "num_tokens": 2904162.0,
      "step": 1670
    },
    {
      "entropy": 6.356344509124756,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.1758,
      "mean_token_accuracy": 0.13067347109317778,
      "num_tokens": 2912783.0,
      "step": 1675
    },
    {
      "entropy": 6.319488096237182,
      "epoch": 0.13071386889710174,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999685206989349,
      "loss": 6.0558,
      "mean_token_accuracy": 0.13287809044122695,
      "num_tokens": 2921027.0,
      "step": 1680
    },
    {
      "entropy": 6.344622087478638,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.093,
      "mean_token_accuracy": 0.13244052454829217,
      "num_tokens": 2930398.0,
      "step": 1685
    },
    {
      "entropy": 6.316963481903076,
      "epoch": 0.13149192764053685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.1126,
      "mean_token_accuracy": 0.1291515350341797,
      "num_tokens": 2938703.0,
      "step": 1690
    },
    {
      "entropy": 6.272860383987426,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.0394,
      "mean_token_accuracy": 0.13433808013796805,
      "num_tokens": 2947007.0,
      "step": 1695
    },
    {
      "entropy": 6.319776296615601,
      "epoch": 0.132269986383972,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.1923,
      "mean_token_accuracy": 0.13449379578232765,
      "num_tokens": 2955485.0,
      "step": 1700
    },
    {
      "entropy": 6.353970575332641,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.0481,
      "mean_token_accuracy": 0.1327715963125229,
      "num_tokens": 2964416.0,
      "step": 1705
    },
    {
      "entropy": 6.239326429367066,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999656776442615,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1316245622932911,
      "num_tokens": 2972737.0,
      "step": 1710
    },
    {
      "entropy": 6.29252290725708,
      "epoch": 0.13343707449912467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.084,
      "mean_token_accuracy": 0.13347245752811432,
      "num_tokens": 2981166.0,
      "step": 1715
    },
    {
      "entropy": 6.313875722885132,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.124,
      "mean_token_accuracy": 0.13230146765708922,
      "num_tokens": 2989689.0,
      "step": 1720
    },
    {
      "entropy": 6.354332733154297,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049996421003463,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1331894539296627,
      "num_tokens": 2998793.0,
      "step": 1725
    },
    {
      "entropy": 6.235461759567261,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.1931,
      "mean_token_accuracy": 0.12982380986213685,
      "num_tokens": 3007320.0,
      "step": 1730
    },
    {
      "entropy": 6.328571414947509,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.0911,
      "mean_token_accuracy": 0.13207585886120796,
      "num_tokens": 3015024.0,
      "step": 1735
    },
    {
      "entropy": 6.351219606399536,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999627117037343,
      "loss": 6.1831,
      "mean_token_accuracy": 0.1294490687549114,
      "num_tokens": 3023058.0,
      "step": 1740
    },
    {
      "entropy": 6.363010501861572,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049996220543319,
      "loss": 6.0924,
      "mean_token_accuracy": 0.13458985537290574,
      "num_tokens": 3031308.0,
      "step": 1745
    },
    {
      "entropy": 6.337850189208984,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.1513,
      "mean_token_accuracy": 0.1289216734468937,
      "num_tokens": 3039598.0,
      "step": 1750
    },
    {
      "entropy": 6.253518009185791,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999611826517793,
      "loss": 5.9714,
      "mean_token_accuracy": 0.1398603707551956,
      "num_tokens": 3047598.0,
      "step": 1755
    },
    {
      "entropy": 6.273720216751099,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.1199,
      "mean_token_accuracy": 0.1361253760755062,
      "num_tokens": 3055702.0,
      "step": 1760
    },
    {
      "entropy": 6.330643892288208,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999601462166569,
      "loss": 5.9841,
      "mean_token_accuracy": 0.14291464164853096,
      "num_tokens": 3063908.0,
      "step": 1765
    },
    {
      "entropy": 6.251221227645874,
      "epoch": 0.1377163975880179,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.0754,
      "mean_token_accuracy": 0.1333239495754242,
      "num_tokens": 3072077.0,
      "step": 1770
    },
    {
      "entropy": 6.303908920288086,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999590961278859,
      "loss": 6.0791,
      "mean_token_accuracy": 0.13312057927250862,
      "num_tokens": 3080498.0,
      "step": 1775
    },
    {
      "entropy": 6.31710524559021,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999585659634021,
      "loss": 6.0814,
      "mean_token_accuracy": 0.1339930735528469,
      "num_tokens": 3089055.0,
      "step": 1780
    },
    {
      "entropy": 6.248744344711303,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999580323855301,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1378532275557518,
      "num_tokens": 3096944.0,
      "step": 1785
    },
    {
      "entropy": 6.228180599212647,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999574953942781,
      "loss": 5.965,
      "mean_token_accuracy": 0.13897001892328262,
      "num_tokens": 3105596.0,
      "step": 1790
    },
    {
      "entropy": 6.191677761077881,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999569549896541,
      "loss": 5.958,
      "mean_token_accuracy": 0.14424095302820206,
      "num_tokens": 3114349.0,
      "step": 1795
    },
    {
      "entropy": 6.258058547973633,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.0799,
      "mean_token_accuracy": 0.12551943436264992,
      "num_tokens": 3122490.0,
      "step": 1800
    },
    {
      "entropy": 6.355373287200928,
      "epoch": 0.14043960319004084,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.1203,
      "mean_token_accuracy": 0.13192636370658875,
      "num_tokens": 3130988.0,
      "step": 1805
    },
    {
      "entropy": 6.246408748626709,
      "epoch": 0.1408286325617584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999553132956328,
      "loss": 5.9784,
      "mean_token_accuracy": 0.13481251522898674,
      "num_tokens": 3139488.0,
      "step": 1810
    },
    {
      "entropy": 6.221766948699951,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999547592376035,
      "loss": 6.0288,
      "mean_token_accuracy": 0.13629044219851494,
      "num_tokens": 3147955.0,
      "step": 1815
    },
    {
      "entropy": 6.249310064315796,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999542017662438,
      "loss": 6.0759,
      "mean_token_accuracy": 0.1276159830391407,
      "num_tokens": 3157608.0,
      "step": 1820
    },
    {
      "entropy": 6.328677940368652,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.0303,
      "mean_token_accuracy": 0.130377709120512,
      "num_tokens": 3166124.0,
      "step": 1825
    },
    {
      "entropy": 6.159740209579468,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999530765835672,
      "loss": 6.0358,
      "mean_token_accuracy": 0.13254115357995033,
      "num_tokens": 3175069.0,
      "step": 1830
    },
    {
      "entropy": 6.292894077301026,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.0664,
      "mean_token_accuracy": 0.13529409915208818,
      "num_tokens": 3184259.0,
      "step": 1835
    },
    {
      "entropy": 6.275870656967163,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999519377476711,
      "loss": 5.9672,
      "mean_token_accuracy": 0.1381848506629467,
      "num_tokens": 3193015.0,
      "step": 1840
    },
    {
      "entropy": 6.267138957977295,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999513632097872,
      "loss": 5.9147,
      "mean_token_accuracy": 0.14502743408083915,
      "num_tokens": 3201624.0,
      "step": 1845
    },
    {
      "entropy": 6.208207750320435,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999507852586245,
      "loss": 6.0294,
      "mean_token_accuracy": 0.13966942057013512,
      "num_tokens": 3209474.0,
      "step": 1850
    },
    {
      "entropy": 6.183934068679809,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.0023,
      "mean_token_accuracy": 0.13654958382248877,
      "num_tokens": 3218550.0,
      "step": 1855
    },
    {
      "entropy": 6.118973255157471,
      "epoch": 0.14471892627893407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999496191164975,
      "loss": 5.9653,
      "mean_token_accuracy": 0.13626407235860824,
      "num_tokens": 3227108.0,
      "step": 1860
    },
    {
      "entropy": 6.250623989105224,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499949030925551,
      "loss": 6.0032,
      "mean_token_accuracy": 0.13892803117632865,
      "num_tokens": 3235785.0,
      "step": 1865
    },
    {
      "entropy": 6.293147802352905,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999484393213609,
      "loss": 6.0377,
      "mean_token_accuracy": 0.1293921411037445,
      "num_tokens": 3244180.0,
      "step": 1870
    },
    {
      "entropy": 6.158266830444336,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999478443039363,
      "loss": 5.9157,
      "mean_token_accuracy": 0.14264118522405625,
      "num_tokens": 3253098.0,
      "step": 1875
    },
    {
      "entropy": 6.26281852722168,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999472458732862,
      "loss": 5.962,
      "mean_token_accuracy": 0.14164620488882065,
      "num_tokens": 3261726.0,
      "step": 1880
    },
    {
      "entropy": 6.122699928283692,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999466440294198,
      "loss": 5.9218,
      "mean_token_accuracy": 0.13797034472227096,
      "num_tokens": 3270362.0,
      "step": 1885
    },
    {
      "entropy": 6.218784666061401,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499946038772346,
      "loss": 5.9119,
      "mean_token_accuracy": 0.1409492813050747,
      "num_tokens": 3278332.0,
      "step": 1890
    },
    {
      "entropy": 6.106905126571656,
      "epoch": 0.147442131880957,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999454301020741,
      "loss": 5.8984,
      "mean_token_accuracy": 0.14537974447011948,
      "num_tokens": 3286800.0,
      "step": 1895
    },
    {
      "entropy": 6.261422252655029,
      "epoch": 0.14783116125267456,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999448180186133,
      "loss": 6.1243,
      "mean_token_accuracy": 0.12735687866806983,
      "num_tokens": 3296186.0,
      "step": 1900
    },
    {
      "entropy": 6.170867204666138,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499944202521973,
      "loss": 5.9874,
      "mean_token_accuracy": 0.14373966082930564,
      "num_tokens": 3303973.0,
      "step": 1905
    },
    {
      "entropy": 6.154075765609742,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999435836121624,
      "loss": 5.9417,
      "mean_token_accuracy": 0.13721608817577363,
      "num_tokens": 3312550.0,
      "step": 1910
    },
    {
      "entropy": 6.231463432312012,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499942961289191,
      "loss": 6.0136,
      "mean_token_accuracy": 0.13601985946297646,
      "num_tokens": 3321342.0,
      "step": 1915
    },
    {
      "entropy": 6.136085319519043,
      "epoch": 0.14938727873954483,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999423355530683,
      "loss": 5.8833,
      "mean_token_accuracy": 0.14982874765992166,
      "num_tokens": 3329293.0,
      "step": 1920
    },
    {
      "entropy": 6.077541828155518,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999417064038035,
      "loss": 5.947,
      "mean_token_accuracy": 0.135797318816185,
      "num_tokens": 3338278.0,
      "step": 1925
    },
    {
      "entropy": 6.248018217086792,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999410738414065,
      "loss": 6.0446,
      "mean_token_accuracy": 0.1355997994542122,
      "num_tokens": 3346757.0,
      "step": 1930
    },
    {
      "entropy": 6.1532470226287845,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999404378658866,
      "loss": 6.0451,
      "mean_token_accuracy": 0.1347198210656643,
      "num_tokens": 3355903.0,
      "step": 1935
    },
    {
      "entropy": 6.208436155319214,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999397984772536,
      "loss": 6.0141,
      "mean_token_accuracy": 0.12824626341462136,
      "num_tokens": 3365440.0,
      "step": 1940
    },
    {
      "entropy": 6.216444635391236,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999391556755172,
      "loss": 5.9763,
      "mean_token_accuracy": 0.1436944492161274,
      "num_tokens": 3373672.0,
      "step": 1945
    },
    {
      "entropy": 6.172102642059326,
      "epoch": 0.15172145496985023,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999385094606872,
      "loss": 5.9552,
      "mean_token_accuracy": 0.13290092200040818,
      "num_tokens": 3382733.0,
      "step": 1950
    },
    {
      "entropy": 6.100259685516358,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999378598327733,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1429830349981785,
      "num_tokens": 3390842.0,
      "step": 1955
    },
    {
      "entropy": 6.200795269012451,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999372067917854,
      "loss": 6.061,
      "mean_token_accuracy": 0.13028706163167952,
      "num_tokens": 3399761.0,
      "step": 1960
    },
    {
      "entropy": 6.219650220870972,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999365503377335,
      "loss": 5.9837,
      "mean_token_accuracy": 0.13638316243886947,
      "num_tokens": 3408986.0,
      "step": 1965
    },
    {
      "entropy": 6.199184322357178,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999358904706273,
      "loss": 5.9356,
      "mean_token_accuracy": 0.13719403147697448,
      "num_tokens": 3418051.0,
      "step": 1970
    },
    {
      "entropy": 6.225965929031372,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999352271904771,
      "loss": 6.0436,
      "mean_token_accuracy": 0.13756765872240068,
      "num_tokens": 3427185.0,
      "step": 1975
    },
    {
      "entropy": 6.179631471633911,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999345604972928,
      "loss": 5.9495,
      "mean_token_accuracy": 0.13750605583190917,
      "num_tokens": 3435367.0,
      "step": 1980
    },
    {
      "entropy": 6.154869127273559,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999338903910846,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1389468789100647,
      "num_tokens": 3443451.0,
      "step": 1985
    },
    {
      "entropy": 6.200529718399048,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999332168718626,
      "loss": 5.9893,
      "mean_token_accuracy": 0.1442691780626774,
      "num_tokens": 3451038.0,
      "step": 1990
    },
    {
      "entropy": 6.408123874664307,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999325399396371,
      "loss": 6.1241,
      "mean_token_accuracy": 0.1256192870438099,
      "num_tokens": 3459693.0,
      "step": 1995
    },
    {
      "entropy": 6.269746255874634,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999318595944183,
      "loss": 6.0106,
      "mean_token_accuracy": 0.13469778150320053,
      "num_tokens": 3468347.0,
      "step": 2000
    },
    {
      "entropy": 6.145987224578858,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999311758362166,
      "loss": 5.986,
      "mean_token_accuracy": 0.13514223098754882,
      "num_tokens": 3476030.0,
      "step": 2005
    },
    {
      "entropy": 6.12372317314148,
      "epoch": 0.156389807430461,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999304886650422,
      "loss": 5.9609,
      "mean_token_accuracy": 0.13282474353909493,
      "num_tokens": 3484840.0,
      "step": 2010
    },
    {
      "entropy": 6.290483140945435,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999297980809058,
      "loss": 6.0489,
      "mean_token_accuracy": 0.13376422077417374,
      "num_tokens": 3492662.0,
      "step": 2015
    },
    {
      "entropy": 6.2352056980133055,
      "epoch": 0.15716786617389614,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999291040838176,
      "loss": 5.9688,
      "mean_token_accuracy": 0.13237249329686165,
      "num_tokens": 3501582.0,
      "step": 2020
    },
    {
      "entropy": 6.179763507843018,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999284066737884,
      "loss": 5.9639,
      "mean_token_accuracy": 0.14239997863769532,
      "num_tokens": 3510141.0,
      "step": 2025
    },
    {
      "entropy": 6.209929418563843,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999277058508285,
      "loss": 6.0049,
      "mean_token_accuracy": 0.14213063940405846,
      "num_tokens": 3518530.0,
      "step": 2030
    },
    {
      "entropy": 6.1083790302276615,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999270016149486,
      "loss": 5.9775,
      "mean_token_accuracy": 0.1358012966811657,
      "num_tokens": 3526815.0,
      "step": 2035
    },
    {
      "entropy": 6.179563379287719,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999262939661596,
      "loss": 6.0282,
      "mean_token_accuracy": 0.13272086828947066,
      "num_tokens": 3535840.0,
      "step": 2040
    },
    {
      "entropy": 6.240152835845947,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999255829044721,
      "loss": 6.0579,
      "mean_token_accuracy": 0.13129353299736976,
      "num_tokens": 3544876.0,
      "step": 2045
    },
    {
      "entropy": 6.140031433105468,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999248684298968,
      "loss": 5.7949,
      "mean_token_accuracy": 0.14725889563560485,
      "num_tokens": 3553018.0,
      "step": 2050
    },
    {
      "entropy": 6.0304547309875485,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999241505424447,
      "loss": 5.8511,
      "mean_token_accuracy": 0.1435144364833832,
      "num_tokens": 3560909.0,
      "step": 2055
    },
    {
      "entropy": 6.116934061050415,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999234292421265,
      "loss": 5.9999,
      "mean_token_accuracy": 0.1397472143173218,
      "num_tokens": 3569235.0,
      "step": 2060
    },
    {
      "entropy": 6.105283117294311,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999227045289535,
      "loss": 5.8792,
      "mean_token_accuracy": 0.14063235372304916,
      "num_tokens": 3577646.0,
      "step": 2065
    },
    {
      "entropy": 6.068844556808472,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999219764029363,
      "loss": 5.9745,
      "mean_token_accuracy": 0.1359837457537651,
      "num_tokens": 3586499.0,
      "step": 2070
    },
    {
      "entropy": 6.221855163574219,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.9812,
      "mean_token_accuracy": 0.14166299402713775,
      "num_tokens": 3595391.0,
      "step": 2075
    },
    {
      "entropy": 6.117988729476929,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999205099124141,
      "loss": 5.903,
      "mean_token_accuracy": 0.14056897163391113,
      "num_tokens": 3603774.0,
      "step": 2080
    },
    {
      "entropy": 6.214944982528687,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999197715479313,
      "loss": 5.9841,
      "mean_token_accuracy": 0.13269646242260932,
      "num_tokens": 3611980.0,
      "step": 2085
    },
    {
      "entropy": 6.1437524318695065,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999190297706489,
      "loss": 5.902,
      "mean_token_accuracy": 0.1393118217587471,
      "num_tokens": 3620902.0,
      "step": 2090
    },
    {
      "entropy": 6.299076747894287,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999182845805783,
      "loss": 6.1871,
      "mean_token_accuracy": 0.12861791327595712,
      "num_tokens": 3628924.0,
      "step": 2095
    },
    {
      "entropy": 6.2256190299987795,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999175359777306,
      "loss": 6.0231,
      "mean_token_accuracy": 0.13702933117747307,
      "num_tokens": 3637538.0,
      "step": 2100
    },
    {
      "entropy": 6.148886060714721,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999167839621174,
      "loss": 6.0424,
      "mean_token_accuracy": 0.130807376652956,
      "num_tokens": 3646309.0,
      "step": 2105
    },
    {
      "entropy": 6.18635630607605,
      "epoch": 0.1641703948648123,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999160285337501,
      "loss": 5.8807,
      "mean_token_accuracy": 0.14085181877017022,
      "num_tokens": 3655204.0,
      "step": 2110
    },
    {
      "entropy": 6.139962005615234,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999152696926399,
      "loss": 6.031,
      "mean_token_accuracy": 0.1334392912685871,
      "num_tokens": 3663939.0,
      "step": 2115
    },
    {
      "entropy": 6.174469661712647,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999145074387985,
      "loss": 5.9681,
      "mean_token_accuracy": 0.13877132534980774,
      "num_tokens": 3673167.0,
      "step": 2120
    },
    {
      "entropy": 6.170883893966675,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999137417722374,
      "loss": 5.9079,
      "mean_token_accuracy": 0.1417030766606331,
      "num_tokens": 3682234.0,
      "step": 2125
    },
    {
      "entropy": 6.110505771636963,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999129726929684,
      "loss": 5.904,
      "mean_token_accuracy": 0.14176712557673454,
      "num_tokens": 3690835.0,
      "step": 2130
    },
    {
      "entropy": 6.160515642166137,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999122002010029,
      "loss": 5.9654,
      "mean_token_accuracy": 0.13802013844251632,
      "num_tokens": 3699348.0,
      "step": 2135
    },
    {
      "entropy": 6.086480808258057,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999114242963527,
      "loss": 5.8711,
      "mean_token_accuracy": 0.1402495212852955,
      "num_tokens": 3708205.0,
      "step": 2140
    },
    {
      "entropy": 6.106543445587159,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999106449790298,
      "loss": 5.9423,
      "mean_token_accuracy": 0.1402938574552536,
      "num_tokens": 3717016.0,
      "step": 2145
    },
    {
      "entropy": 6.246592807769775,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999098622490458,
      "loss": 5.9544,
      "mean_token_accuracy": 0.13701125010848045,
      "num_tokens": 3725509.0,
      "step": 2150
    },
    {
      "entropy": 6.154575967788697,
      "epoch": 0.16767165921027039,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999090761064125,
      "loss": 6.0286,
      "mean_token_accuracy": 0.1339248612523079,
      "num_tokens": 3734674.0,
      "step": 2155
    },
    {
      "entropy": 6.060240173339844,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499908286551142,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1407431736588478,
      "num_tokens": 3742930.0,
      "step": 2160
    },
    {
      "entropy": 6.154189491271973,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999074935832463,
      "loss": 5.9369,
      "mean_token_accuracy": 0.1409689135849476,
      "num_tokens": 3751314.0,
      "step": 2165
    },
    {
      "entropy": 6.216179037094117,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999066972027373,
      "loss": 6.0535,
      "mean_token_accuracy": 0.13319066762924195,
      "num_tokens": 3760229.0,
      "step": 2170
    },
    {
      "entropy": 6.285803413391113,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999058974096271,
      "loss": 6.0545,
      "mean_token_accuracy": 0.1335298717021942,
      "num_tokens": 3769037.0,
      "step": 2175
    },
    {
      "entropy": 6.052758312225341,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499905094203928,
      "loss": 5.9238,
      "mean_token_accuracy": 0.13934388458728791,
      "num_tokens": 3777686.0,
      "step": 2180
    },
    {
      "entropy": 6.218620347976684,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499904287585652,
      "loss": 6.0044,
      "mean_token_accuracy": 0.13825156837701796,
      "num_tokens": 3786412.0,
      "step": 2185
    },
    {
      "entropy": 6.192234373092651,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999034775548113,
      "loss": 6.0351,
      "mean_token_accuracy": 0.13853697702288628,
      "num_tokens": 3794981.0,
      "step": 2190
    },
    {
      "entropy": 6.155090808868408,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.9303,
      "mean_token_accuracy": 0.14125299602746963,
      "num_tokens": 3803069.0,
      "step": 2195
    },
    {
      "entropy": 6.151627492904663,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999018472554857,
      "loss": 6.0112,
      "mean_token_accuracy": 0.1394211769104004,
      "num_tokens": 3811986.0,
      "step": 2200
    },
    {
      "entropy": 6.219188404083252,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999010269870253,
      "loss": 5.9302,
      "mean_token_accuracy": 0.13892962485551835,
      "num_tokens": 3820626.0,
      "step": 2205
    },
    {
      "entropy": 6.105683422088623,
      "epoch": 0.17195098229916359,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999002033060498,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1396380692720413,
      "num_tokens": 3828608.0,
      "step": 2210
    },
    {
      "entropy": 6.041509485244751,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998993762125716,
      "loss": 5.8795,
      "mean_token_accuracy": 0.1437774755060673,
      "num_tokens": 3837059.0,
      "step": 2215
    },
    {
      "entropy": 6.148551082611084,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998985457066034,
      "loss": 5.8953,
      "mean_token_accuracy": 0.13735355138778688,
      "num_tokens": 3845137.0,
      "step": 2220
    },
    {
      "entropy": 6.128031158447266,
      "epoch": 0.1731180704143163,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998977117881576,
      "loss": 5.8861,
      "mean_token_accuracy": 0.14574987590312957,
      "num_tokens": 3854012.0,
      "step": 2225
    },
    {
      "entropy": 6.015646696090698,
      "epoch": 0.17350709978603385,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998968744572472,
      "loss": 5.8295,
      "mean_token_accuracy": 0.1409018814563751,
      "num_tokens": 3862308.0,
      "step": 2230
    },
    {
      "entropy": 6.067210721969604,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.7717,
      "mean_token_accuracy": 0.13879043981432915,
      "num_tokens": 3870975.0,
      "step": 2235
    },
    {
      "entropy": 6.087196779251099,
      "epoch": 0.17428515852946896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998951895580826,
      "loss": 5.9544,
      "mean_token_accuracy": 0.14272485896945,
      "num_tokens": 3879621.0,
      "step": 2240
    },
    {
      "entropy": 6.093369865417481,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.9166,
      "mean_token_accuracy": 0.13817212730646133,
      "num_tokens": 3889214.0,
      "step": 2245
    },
    {
      "entropy": 6.146738815307617,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.9504,
      "mean_token_accuracy": 0.14284390732645988,
      "num_tokens": 3898250.0,
      "step": 2250
    },
    {
      "entropy": 6.129539108276367,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998926366161689,
      "loss": 5.9798,
      "mean_token_accuracy": 0.13206198140978814,
      "num_tokens": 3906971.0,
      "step": 2255
    },
    {
      "entropy": 6.1653541088104244,
      "epoch": 0.17584127601633923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998917788107383,
      "loss": 6.0139,
      "mean_token_accuracy": 0.13972242698073387,
      "num_tokens": 3916663.0,
      "step": 2260
    },
    {
      "entropy": 6.103351974487305,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998909175929326,
      "loss": 5.8942,
      "mean_token_accuracy": 0.14154435321688652,
      "num_tokens": 3926848.0,
      "step": 2265
    },
    {
      "entropy": 6.034053564071655,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998900529627653,
      "loss": 5.8613,
      "mean_token_accuracy": 0.1404354229569435,
      "num_tokens": 3935729.0,
      "step": 2270
    },
    {
      "entropy": 6.141492509841919,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.8766,
      "mean_token_accuracy": 0.1424740418791771,
      "num_tokens": 3943408.0,
      "step": 2275
    },
    {
      "entropy": 6.052723979949951,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499888313465398,
      "loss": 5.8509,
      "mean_token_accuracy": 0.14049414843320845,
      "num_tokens": 3952473.0,
      "step": 2280
    },
    {
      "entropy": 6.178114128112793,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998874385982247,
      "loss": 6.0018,
      "mean_token_accuracy": 0.13325022533535957,
      "num_tokens": 3960878.0,
      "step": 2285
    },
    {
      "entropy": 6.208947801589966,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.945,
      "mean_token_accuracy": 0.13925340324640273,
      "num_tokens": 3969282.0,
      "step": 2290
    },
    {
      "entropy": 6.132673740386963,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499885678626964,
      "loss": 5.9267,
      "mean_token_accuracy": 0.14256497994065284,
      "num_tokens": 3978516.0,
      "step": 2295
    },
    {
      "entropy": 6.066269874572754,
      "epoch": 0.17895351099007975,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998847935229038,
      "loss": 5.7855,
      "mean_token_accuracy": 0.15060577318072318,
      "num_tokens": 3986580.0,
      "step": 2300
    },
    {
      "entropy": 5.910192060470581,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998839050065746,
      "loss": 5.8052,
      "mean_token_accuracy": 0.14743324518203735,
      "num_tokens": 3995144.0,
      "step": 2305
    },
    {
      "entropy": 6.101610565185547,
      "epoch": 0.17973156973351487,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998830130779902,
      "loss": 5.8075,
      "mean_token_accuracy": 0.13889141082763673,
      "num_tokens": 4003887.0,
      "step": 2310
    },
    {
      "entropy": 6.0803686618804935,
      "epoch": 0.18012059910523245,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499882117737164,
      "loss": 5.9209,
      "mean_token_accuracy": 0.14422064125537873,
      "num_tokens": 4012607.0,
      "step": 2315
    },
    {
      "entropy": 6.053954172134399,
      "epoch": 0.18050962847695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998812189841096,
      "loss": 6.0034,
      "mean_token_accuracy": 0.13622427582740784,
      "num_tokens": 4021950.0,
      "step": 2320
    },
    {
      "entropy": 6.1665068626403805,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998803168188407,
      "loss": 5.9668,
      "mean_token_accuracy": 0.1415689304471016,
      "num_tokens": 4031076.0,
      "step": 2325
    },
    {
      "entropy": 5.990097284317017,
      "epoch": 0.18128768722038513,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.8054,
      "mean_token_accuracy": 0.14801988750696182,
      "num_tokens": 4040414.0,
      "step": 2330
    },
    {
      "entropy": 6.18867506980896,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998785022517137,
      "loss": 5.8924,
      "mean_token_accuracy": 0.14523087441921234,
      "num_tokens": 4048931.0,
      "step": 2335
    },
    {
      "entropy": 5.915593385696411,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.7899,
      "mean_token_accuracy": 0.1424514263868332,
      "num_tokens": 4057688.0,
      "step": 2340
    },
    {
      "entropy": 6.047796154022217,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998766740358936,
      "loss": 5.9117,
      "mean_token_accuracy": 0.14413589462637902,
      "num_tokens": 4066365.0,
      "step": 2345
    },
    {
      "entropy": 6.0719398021697994,
      "epoch": 0.1828438047072554,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998757548097582,
      "loss": 5.7968,
      "mean_token_accuracy": 0.14763821363449098,
      "num_tokens": 4075136.0,
      "step": 2350
    },
    {
      "entropy": 6.112574434280395,
      "epoch": 0.18323283407897295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998748321714911,
      "loss": 5.9058,
      "mean_token_accuracy": 0.1457724839448929,
      "num_tokens": 4083337.0,
      "step": 2355
    },
    {
      "entropy": 6.00722303390503,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998739061211065,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14413827881217003,
      "num_tokens": 4091201.0,
      "step": 2360
    },
    {
      "entropy": 6.140283298492432,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.9072,
      "mean_token_accuracy": 0.14183114990592002,
      "num_tokens": 4099480.0,
      "step": 2365
    },
    {
      "entropy": 5.984076595306396,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998720437840403,
      "loss": 5.8158,
      "mean_token_accuracy": 0.1437339261174202,
      "num_tokens": 4108151.0,
      "step": 2370
    },
    {
      "entropy": 5.961683034896851,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.8251,
      "mean_token_accuracy": 0.1457428365945816,
      "num_tokens": 4117380.0,
      "step": 2375
    },
    {
      "entropy": 6.036372423171997,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.8689,
      "mean_token_accuracy": 0.14735556095838548,
      "num_tokens": 4126757.0,
      "step": 2380
    },
    {
      "entropy": 6.031168508529663,
      "epoch": 0.18556701030927836,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998692246879116,
      "loss": 5.7891,
      "mean_token_accuracy": 0.146223184466362,
      "num_tokens": 4135326.0,
      "step": 2385
    },
    {
      "entropy": 6.053634834289551,
      "epoch": 0.18595603968099592,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.7988,
      "mean_token_accuracy": 0.15181397199630736,
      "num_tokens": 4143624.0,
      "step": 2390
    },
    {
      "entropy": 5.992155838012695,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.7659,
      "mean_token_accuracy": 0.14670938700437547,
      "num_tokens": 4152193.0,
      "step": 2395
    },
    {
      "entropy": 6.03981647491455,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.8835,
      "mean_token_accuracy": 0.141094571352005,
      "num_tokens": 4160614.0,
      "step": 2400
    },
    {
      "entropy": 6.138073205947876,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998654181246851,
      "loss": 5.9151,
      "mean_token_accuracy": 0.1435561805963516,
      "num_tokens": 4168715.0,
      "step": 2405
    },
    {
      "entropy": 6.1046959400177006,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998644579539052,
      "loss": 5.956,
      "mean_token_accuracy": 0.1427804246544838,
      "num_tokens": 4176676.0,
      "step": 2410
    },
    {
      "entropy": 6.041570472717285,
      "epoch": 0.18790118653958374,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499863494371165,
      "loss": 5.9196,
      "mean_token_accuracy": 0.13990830332040788,
      "num_tokens": 4185613.0,
      "step": 2415
    },
    {
      "entropy": 6.056922721862793,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998625273764793,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1435866177082062,
      "num_tokens": 4193985.0,
      "step": 2420
    },
    {
      "entropy": 6.057398271560669,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.8347,
      "mean_token_accuracy": 0.1472941130399704,
      "num_tokens": 4202713.0,
      "step": 2425
    },
    {
      "entropy": 6.15912766456604,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.9598,
      "mean_token_accuracy": 0.13513548597693442,
      "num_tokens": 4210858.0,
      "step": 2430
    },
    {
      "entropy": 6.027268743515014,
      "epoch": 0.189457304026454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.876,
      "mean_token_accuracy": 0.14263733327388764,
      "num_tokens": 4219570.0,
      "step": 2435
    },
    {
      "entropy": 6.034002113342285,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998586252785743,
      "loss": 5.7873,
      "mean_token_accuracy": 0.14603268951177598,
      "num_tokens": 4228026.0,
      "step": 2440
    },
    {
      "entropy": 6.043663167953492,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.8084,
      "mean_token_accuracy": 0.14466246217489243,
      "num_tokens": 4235993.0,
      "step": 2445
    },
    {
      "entropy": 6.009197854995728,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.8151,
      "mean_token_accuracy": 0.147155824303627,
      "num_tokens": 4244225.0,
      "step": 2450
    },
    {
      "entropy": 5.965597009658813,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.7885,
      "mean_token_accuracy": 0.1485578991472721,
      "num_tokens": 4252989.0,
      "step": 2455
    },
    {
      "entropy": 6.069110679626465,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.8298,
      "mean_token_accuracy": 0.14657569378614427,
      "num_tokens": 4261297.0,
      "step": 2460
    },
    {
      "entropy": 6.0928350448608395,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.9644,
      "mean_token_accuracy": 0.13758677765727043,
      "num_tokens": 4270299.0,
      "step": 2465
    },
    {
      "entropy": 6.128661823272705,
      "epoch": 0.19218050962847694,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.9728,
      "mean_token_accuracy": 0.13565355464816092,
      "num_tokens": 4279662.0,
      "step": 2470
    },
    {
      "entropy": 6.051914119720459,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998516652507585,
      "loss": 5.7902,
      "mean_token_accuracy": 0.1471126437187195,
      "num_tokens": 4287876.0,
      "step": 2475
    },
    {
      "entropy": 6.014107513427734,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.9218,
      "mean_token_accuracy": 0.14491380155086517,
      "num_tokens": 4296415.0,
      "step": 2480
    },
    {
      "entropy": 5.982778406143188,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.7484,
      "mean_token_accuracy": 0.14843220636248589,
      "num_tokens": 4304434.0,
      "step": 2485
    },
    {
      "entropy": 6.051380062103272,
      "epoch": 0.1937366271153472,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998486312048995,
      "loss": 5.9265,
      "mean_token_accuracy": 0.1381976492702961,
      "num_tokens": 4313266.0,
      "step": 2490
    },
    {
      "entropy": 6.0365697860717775,
      "epoch": 0.19412565648706478,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.8013,
      "mean_token_accuracy": 0.14801377803087234,
      "num_tokens": 4322121.0,
      "step": 2495
    },
    {
      "entropy": 5.980331611633301,
      "epoch": 0.19451468585878234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998465914490343,
      "loss": 5.8202,
      "mean_token_accuracy": 0.14618535935878754,
      "num_tokens": 4330596.0,
      "step": 2500
    },
    {
      "entropy": 6.062447881698608,
      "epoch": 0.1949037152304999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049984556645356,
      "loss": 5.9022,
      "mean_token_accuracy": 0.1466861292719841,
      "num_tokens": 4337779.0,
      "step": 2505
    },
    {
      "entropy": 6.0627391815185545,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.7454,
      "mean_token_accuracy": 0.14693358018994332,
      "num_tokens": 4346623.0,
      "step": 2510
    },
    {
      "entropy": 5.929016637802124,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.8526,
      "mean_token_accuracy": 0.14088083803653717,
      "num_tokens": 4355435.0,
      "step": 2515
    },
    {
      "entropy": 6.1320250034332275,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998424709971586,
      "loss": 5.902,
      "mean_token_accuracy": 0.14303931146860122,
      "num_tokens": 4363636.0,
      "step": 2520
    },
    {
      "entropy": 5.970055866241455,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.7269,
      "mean_token_accuracy": 0.15075851678848268,
      "num_tokens": 4372337.0,
      "step": 2525
    },
    {
      "entropy": 5.979619741439819,
      "epoch": 0.19684886208908772,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998403903013984,
      "loss": 5.7561,
      "mean_token_accuracy": 0.15269921571016312,
      "num_tokens": 4380528.0,
      "step": 2530
    },
    {
      "entropy": 5.998526477813721,
      "epoch": 0.19723789146080528,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.8501,
      "mean_token_accuracy": 0.13916658014059066,
      "num_tokens": 4389464.0,
      "step": 2535
    },
    {
      "entropy": 6.082704496383667,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499838295959258,
      "loss": 5.9085,
      "mean_token_accuracy": 0.14364177733659744,
      "num_tokens": 4397879.0,
      "step": 2540
    },
    {
      "entropy": 6.062457323074341,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499837243670835,
      "loss": 5.932,
      "mean_token_accuracy": 0.14348311796784402,
      "num_tokens": 4406145.0,
      "step": 2545
    },
    {
      "entropy": 6.0163922786712645,
      "epoch": 0.19840497957595798,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.8382,
      "mean_token_accuracy": 0.14055323377251625,
      "num_tokens": 4415480.0,
      "step": 2550
    },
    {
      "entropy": 6.038650321960449,
      "epoch": 0.19879400894767554,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499835128859363,
      "loss": 5.8457,
      "mean_token_accuracy": 0.14373011589050294,
      "num_tokens": 4423355.0,
      "step": 2555
    },
    {
      "entropy": 5.971053647994995,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.7288,
      "mean_token_accuracy": 0.15200028717517852,
      "num_tokens": 4432059.0,
      "step": 2560
    },
    {
      "entropy": 5.9389901638031,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.8266,
      "mean_token_accuracy": 0.14117742702364922,
      "num_tokens": 4441146.0,
      "step": 2565
    },
    {
      "entropy": 5.973790550231934,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.6701,
      "mean_token_accuracy": 0.15412732362747192,
      "num_tokens": 4448553.0,
      "step": 2570
    },
    {
      "entropy": 5.836313152313233,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.7375,
      "mean_token_accuracy": 0.15181366503238677,
      "num_tokens": 4456930.0,
      "step": 2575
    },
    {
      "entropy": 6.073364543914795,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.9386,
      "mean_token_accuracy": 0.14404268860816954,
      "num_tokens": 4464932.0,
      "step": 2580
    },
    {
      "entropy": 5.992941284179688,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.6672,
      "mean_token_accuracy": 0.15902086794376374,
      "num_tokens": 4473695.0,
      "step": 2585
    },
    {
      "entropy": 5.925492858886718,
      "epoch": 0.2015172145496985,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.8921,
      "mean_token_accuracy": 0.14221425950527192,
      "num_tokens": 4483375.0,
      "step": 2590
    },
    {
      "entropy": 6.11337890625,
      "epoch": 0.20190624392141607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.9193,
      "mean_token_accuracy": 0.13467642813920974,
      "num_tokens": 4491695.0,
      "step": 2595
    },
    {
      "entropy": 6.003162908554077,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998254433396088,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1440841868519783,
      "num_tokens": 4500361.0,
      "step": 2600
    },
    {
      "entropy": 5.987391805648803,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.8079,
      "mean_token_accuracy": 0.14264530092477798,
      "num_tokens": 4509368.0,
      "step": 2605
    },
    {
      "entropy": 6.039334154129028,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.7889,
      "mean_token_accuracy": 0.14281965047121048,
      "num_tokens": 4518146.0,
      "step": 2610
    },
    {
      "entropy": 6.01565580368042,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.8405,
      "mean_token_accuracy": 0.14281932562589644,
      "num_tokens": 4526128.0,
      "step": 2615
    },
    {
      "entropy": 6.036164903640747,
      "epoch": 0.2038513907800039,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.8847,
      "mean_token_accuracy": 0.14803217351436615,
      "num_tokens": 4534724.0,
      "step": 2620
    },
    {
      "entropy": 5.968258047103882,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.7427,
      "mean_token_accuracy": 0.14365783631801604,
      "num_tokens": 4543725.0,
      "step": 2625
    },
    {
      "entropy": 6.012339067459107,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.8252,
      "mean_token_accuracy": 0.14723241925239564,
      "num_tokens": 4551758.0,
      "step": 2630
    },
    {
      "entropy": 5.894060182571411,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.7079,
      "mean_token_accuracy": 0.15085572972893715,
      "num_tokens": 4559990.0,
      "step": 2635
    },
    {
      "entropy": 5.93325366973877,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.7558,
      "mean_token_accuracy": 0.1438131585717201,
      "num_tokens": 4568651.0,
      "step": 2640
    },
    {
      "entropy": 5.900708532333374,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.6758,
      "mean_token_accuracy": 0.15049925595521926,
      "num_tokens": 4577981.0,
      "step": 2645
    },
    {
      "entropy": 5.955266237258911,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049981435757157,
      "loss": 5.8312,
      "mean_token_accuracy": 0.147544726729393,
      "num_tokens": 4586480.0,
      "step": 2650
    },
    {
      "entropy": 5.980217790603637,
      "epoch": 0.20657459638202685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1525867000222206,
      "num_tokens": 4595554.0,
      "step": 2655
    },
    {
      "entropy": 5.923847293853759,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.7566,
      "mean_token_accuracy": 0.15113697126507758,
      "num_tokens": 4604091.0,
      "step": 2660
    },
    {
      "entropy": 6.023667430877685,
      "epoch": 0.20735265512546197,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499810965321987,
      "loss": 5.8615,
      "mean_token_accuracy": 0.14118051454424857,
      "num_tokens": 4612603.0,
      "step": 2665
    },
    {
      "entropy": 6.032278680801392,
      "epoch": 0.20774168449717953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998098277498059,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1470929801464081,
      "num_tokens": 4621041.0,
      "step": 2670
    },
    {
      "entropy": 5.90060658454895,
      "epoch": 0.2081307138688971,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.7983,
      "mean_token_accuracy": 0.15032607913017274,
      "num_tokens": 4629849.0,
      "step": 2675
    },
    {
      "entropy": 6.054356479644776,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499807542372067,
      "loss": 5.8687,
      "mean_token_accuracy": 0.14306920617818833,
      "num_tokens": 4639426.0,
      "step": 2680
    },
    {
      "entropy": 5.9633255958557125,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998063945665439,
      "loss": 5.7888,
      "mean_token_accuracy": 0.14868860244750975,
      "num_tokens": 4647257.0,
      "step": 2685
    },
    {
      "entropy": 5.958910369873047,
      "epoch": 0.2092978019840498,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998052433499416,
      "loss": 5.7688,
      "mean_token_accuracy": 0.14971182197332383,
      "num_tokens": 4655773.0,
      "step": 2690
    },
    {
      "entropy": 5.995905542373658,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998040887222776,
      "loss": 5.8141,
      "mean_token_accuracy": 0.14208999648690224,
      "num_tokens": 4664292.0,
      "step": 2695
    },
    {
      "entropy": 6.006470203399658,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1457445926964283,
      "num_tokens": 4672774.0,
      "step": 2700
    },
    {
      "entropy": 6.1288737773895265,
      "epoch": 0.2104648900992025,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998017692338344,
      "loss": 5.954,
      "mean_token_accuracy": 0.14017487764358522,
      "num_tokens": 4681855.0,
      "step": 2705
    },
    {
      "entropy": 5.958335494995117,
      "epoch": 0.21085391947092005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.7565,
      "mean_token_accuracy": 0.14842477068305016,
      "num_tokens": 4689771.0,
      "step": 2710
    },
    {
      "entropy": 6.003065395355224,
      "epoch": 0.2112429488426376,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1353430561721325,
      "num_tokens": 4698808.0,
      "step": 2715
    },
    {
      "entropy": 5.980752849578858,
      "epoch": 0.21163197821435517,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.7782,
      "mean_token_accuracy": 0.14798913151025772,
      "num_tokens": 4707627.0,
      "step": 2720
    },
    {
      "entropy": 5.949896383285522,
      "epoch": 0.21202100758607276,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.7576,
      "mean_token_accuracy": 0.14555923044681549,
      "num_tokens": 4716091.0,
      "step": 2725
    },
    {
      "entropy": 5.944676637649536,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.788,
      "mean_token_accuracy": 0.1488296665251255,
      "num_tokens": 4724042.0,
      "step": 2730
    },
    {
      "entropy": 5.887519454956054,
      "epoch": 0.21279906632950787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.7058,
      "mean_token_accuracy": 0.15533653944730758,
      "num_tokens": 4731745.0,
      "step": 2735
    },
    {
      "entropy": 5.945332145690918,
      "epoch": 0.21318809570122543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997935435784302,
      "loss": 5.872,
      "mean_token_accuracy": 0.146851022541523,
      "num_tokens": 4740386.0,
      "step": 2740
    },
    {
      "entropy": 6.041448068618775,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1391231372952461,
      "num_tokens": 4749809.0,
      "step": 2745
    },
    {
      "entropy": 6.063974905014038,
      "epoch": 0.21396615444466058,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499791162692945,
      "loss": 5.9768,
      "mean_token_accuracy": 0.1409013733267784,
      "num_tokens": 4759507.0,
      "step": 2750
    },
    {
      "entropy": 5.960179090499878,
      "epoch": 0.21435518381637814,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.7479,
      "mean_token_accuracy": 0.1449030451476574,
      "num_tokens": 4768793.0,
      "step": 2755
    },
    {
      "entropy": 6.029723405838013,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.9008,
      "mean_token_accuracy": 0.14465020522475242,
      "num_tokens": 4777344.0,
      "step": 2760
    },
    {
      "entropy": 6.030321025848389,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.8127,
      "mean_token_accuracy": 0.1441117987036705,
      "num_tokens": 4785934.0,
      "step": 2765
    },
    {
      "entropy": 5.947022724151611,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.7731,
      "mean_token_accuracy": 0.1485394760966301,
      "num_tokens": 4794889.0,
      "step": 2770
    },
    {
      "entropy": 5.944930124282837,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.8881,
      "mean_token_accuracy": 0.14024131819605828,
      "num_tokens": 4803597.0,
      "step": 2775
    },
    {
      "entropy": 5.933191776275635,
      "epoch": 0.21630033067496596,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997839381767178,
      "loss": 5.6613,
      "mean_token_accuracy": 0.15584157705307006,
      "num_tokens": 4812706.0,
      "step": 2780
    },
    {
      "entropy": 5.90777177810669,
      "epoch": 0.21668936004668352,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.7295,
      "mean_token_accuracy": 0.1480869472026825,
      "num_tokens": 4821790.0,
      "step": 2785
    },
    {
      "entropy": 5.968650197982788,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.7899,
      "mean_token_accuracy": 0.1485279232263565,
      "num_tokens": 4830123.0,
      "step": 2790
    },
    {
      "entropy": 5.94366946220398,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.7194,
      "mean_token_accuracy": 0.1537564590573311,
      "num_tokens": 4838066.0,
      "step": 2795
    },
    {
      "entropy": 5.9353564262390135,
      "epoch": 0.21785644816183622,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.7944,
      "mean_token_accuracy": 0.14718551486730574,
      "num_tokens": 4846877.0,
      "step": 2800
    },
    {
      "entropy": 5.991869354248047,
      "epoch": 0.21824547753355378,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.8426,
      "mean_token_accuracy": 0.14051205813884735,
      "num_tokens": 4856960.0,
      "step": 2805
    },
    {
      "entropy": 6.004691219329834,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.8542,
      "mean_token_accuracy": 0.14763535112142562,
      "num_tokens": 4865440.0,
      "step": 2810
    },
    {
      "entropy": 5.893306064605713,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.7417,
      "mean_token_accuracy": 0.1489579737186432,
      "num_tokens": 4874187.0,
      "step": 2815
    },
    {
      "entropy": 6.102828645706177,
      "epoch": 0.21941256564870648,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049977411448776,
      "loss": 5.8831,
      "mean_token_accuracy": 0.14227399230003357,
      "num_tokens": 4883151.0,
      "step": 2820
    },
    {
      "entropy": 5.983131647109985,
      "epoch": 0.21980159502042404,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.7977,
      "mean_token_accuracy": 0.14272094666957855,
      "num_tokens": 4890968.0,
      "step": 2825
    },
    {
      "entropy": 5.879379606246948,
      "epoch": 0.2201906243921416,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.6825,
      "mean_token_accuracy": 0.15600915551185607,
      "num_tokens": 4899389.0,
      "step": 2830
    },
    {
      "entropy": 5.905595302581787,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.7509,
      "mean_token_accuracy": 0.15158967077732086,
      "num_tokens": 4907622.0,
      "step": 2835
    },
    {
      "entropy": 5.943753623962403,
      "epoch": 0.22096868313557674,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499769120788537,
      "loss": 5.7491,
      "mean_token_accuracy": 0.1479824811220169,
      "num_tokens": 4916151.0,
      "step": 2840
    },
    {
      "entropy": 5.892854356765747,
      "epoch": 0.2213577125072943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.6692,
      "mean_token_accuracy": 0.15641618967056276,
      "num_tokens": 4923938.0,
      "step": 2845
    },
    {
      "entropy": 5.87450475692749,
      "epoch": 0.22174674187901186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.7613,
      "mean_token_accuracy": 0.16034080535173417,
      "num_tokens": 4932027.0,
      "step": 2850
    },
    {
      "entropy": 5.947768974304199,
      "epoch": 0.22213577125072942,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.7766,
      "mean_token_accuracy": 0.1560262620449066,
      "num_tokens": 4940229.0,
      "step": 2855
    },
    {
      "entropy": 5.907529211044311,
      "epoch": 0.222524800622447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.8405,
      "mean_token_accuracy": 0.14572847858071328,
      "num_tokens": 4948662.0,
      "step": 2860
    },
    {
      "entropy": 5.958647584915161,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.7854,
      "mean_token_accuracy": 0.1521115556359291,
      "num_tokens": 4956847.0,
      "step": 2865
    },
    {
      "entropy": 5.970917129516602,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.7205,
      "mean_token_accuracy": 0.14754560068249703,
      "num_tokens": 4965139.0,
      "step": 2870
    },
    {
      "entropy": 6.0411519527435305,
      "epoch": 0.22369188873759968,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997602505108603,
      "loss": 5.8952,
      "mean_token_accuracy": 0.14166620448231698,
      "num_tokens": 4973671.0,
      "step": 2875
    },
    {
      "entropy": 5.933946704864502,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.7652,
      "mean_token_accuracy": 0.14397089332342147,
      "num_tokens": 4981970.0,
      "step": 2880
    },
    {
      "entropy": 5.935157251358032,
      "epoch": 0.22446994748103483,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.7921,
      "mean_token_accuracy": 0.14724297374486922,
      "num_tokens": 4990591.0,
      "step": 2885
    },
    {
      "entropy": 5.9578015327453615,
      "epoch": 0.22485897685275238,
      "grad_norm": 1.125,
      "learning_rate": 0.000499756397807213,
      "loss": 5.7954,
      "mean_token_accuracy": 0.14315731078386307,
      "num_tokens": 4999623.0,
      "step": 2890
    },
    {
      "entropy": 6.011578512191773,
      "epoch": 0.22524800622446994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.8545,
      "mean_token_accuracy": 0.1422098271548748,
      "num_tokens": 5008387.0,
      "step": 2895
    },
    {
      "entropy": 5.949732828140259,
      "epoch": 0.2256370355961875,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1461174599826336,
      "num_tokens": 5016788.0,
      "step": 2900
    },
    {
      "entropy": 6.023969125747681,
      "epoch": 0.2260260649679051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.8104,
      "mean_token_accuracy": 0.14314667582511903,
      "num_tokens": 5025508.0,
      "step": 2905
    },
    {
      "entropy": 5.927273511886597,
      "epoch": 0.22641509433962265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049975121312462,
      "loss": 5.7821,
      "mean_token_accuracy": 0.14801414608955382,
      "num_tokens": 5035081.0,
      "step": 2910
    },
    {
      "entropy": 5.879549169540406,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.7173,
      "mean_token_accuracy": 0.15523723959922792,
      "num_tokens": 5043655.0,
      "step": 2915
    },
    {
      "entropy": 5.91579909324646,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.74,
      "mean_token_accuracy": 0.15187966972589492,
      "num_tokens": 5052080.0,
      "step": 2920
    },
    {
      "entropy": 5.888910627365112,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.7076,
      "mean_token_accuracy": 0.15852134227752684,
      "num_tokens": 5061325.0,
      "step": 2925
    },
    {
      "entropy": 5.961522054672241,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.8208,
      "mean_token_accuracy": 0.15095880776643752,
      "num_tokens": 5069461.0,
      "step": 2930
    },
    {
      "entropy": 5.988435316085815,
      "epoch": 0.22836024119821047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.7807,
      "mean_token_accuracy": 0.14658573567867278,
      "num_tokens": 5077747.0,
      "step": 2935
    },
    {
      "entropy": 5.988274765014649,
      "epoch": 0.22874927056992803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.8365,
      "mean_token_accuracy": 0.15124830231070518,
      "num_tokens": 5086300.0,
      "step": 2940
    },
    {
      "entropy": 5.844106388092041,
      "epoch": 0.22913829994164558,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.6151,
      "mean_token_accuracy": 0.15467342734336853,
      "num_tokens": 5094762.0,
      "step": 2945
    },
    {
      "entropy": 5.89040789604187,
      "epoch": 0.22952732931336317,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.7742,
      "mean_token_accuracy": 0.14963481277227403,
      "num_tokens": 5103364.0,
      "step": 2950
    },
    {
      "entropy": 6.023597526550293,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.8799,
      "mean_token_accuracy": 0.14407353550195695,
      "num_tokens": 5111730.0,
      "step": 2955
    },
    {
      "entropy": 5.970983409881592,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.6786,
      "mean_token_accuracy": 0.1486763596534729,
      "num_tokens": 5119878.0,
      "step": 2960
    },
    {
      "entropy": 5.9219104766845705,
      "epoch": 0.23069441742851585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1522405095398426,
      "num_tokens": 5129146.0,
      "step": 2965
    },
    {
      "entropy": 5.892798137664795,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.8065,
      "mean_token_accuracy": 0.14348602443933486,
      "num_tokens": 5137989.0,
      "step": 2970
    },
    {
      "entropy": 6.039399528503418,
      "epoch": 0.231472476171951,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.8268,
      "mean_token_accuracy": 0.14684822633862496,
      "num_tokens": 5146722.0,
      "step": 2975
    },
    {
      "entropy": 6.007096481323242,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.8128,
      "mean_token_accuracy": 0.14832943975925444,
      "num_tokens": 5154902.0,
      "step": 2980
    },
    {
      "entropy": 5.963329792022705,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.827,
      "mean_token_accuracy": 0.14517123103141785,
      "num_tokens": 5163594.0,
      "step": 2985
    },
    {
      "entropy": 5.939759683609009,
      "epoch": 0.23263956428710367,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.7065,
      "mean_token_accuracy": 0.1531409054994583,
      "num_tokens": 5172139.0,
      "step": 2990
    },
    {
      "entropy": 5.937909650802612,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.8315,
      "mean_token_accuracy": 0.15084856152534484,
      "num_tokens": 5180077.0,
      "step": 2995
    },
    {
      "entropy": 5.849994754791259,
      "epoch": 0.2334176230305388,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.6851,
      "mean_token_accuracy": 0.14938049018383026,
      "num_tokens": 5188500.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.659352120260876,
      "eval_loss": 5.808426856994629,
      "eval_mean_token_accuracy": 0.15537977798754946,
      "eval_num_tokens": 5188500.0,
      "eval_runtime": 21.8125,
      "eval_samples_per_second": 1905.236,
      "eval_steps_per_second": 238.166,
      "step": 3000
    },
    {
      "entropy": 5.93622145652771,
      "epoch": 0.23380665240225637,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499725840766324,
      "loss": 5.7657,
      "mean_token_accuracy": 0.14401972368359567,
      "num_tokens": 5197214.0,
      "step": 3005
    },
    {
      "entropy": 6.025457191467285,
      "epoch": 0.23419568177397393,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.8495,
      "mean_token_accuracy": 0.14099710583686828,
      "num_tokens": 5206348.0,
      "step": 3010
    },
    {
      "entropy": 5.808966779708863,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.6655,
      "mean_token_accuracy": 0.14906769394874572,
      "num_tokens": 5215425.0,
      "step": 3015
    },
    {
      "entropy": 6.039634037017822,
      "epoch": 0.23497374051740907,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.891,
      "mean_token_accuracy": 0.13933431655168532,
      "num_tokens": 5224899.0,
      "step": 3020
    },
    {
      "entropy": 5.964243030548095,
      "epoch": 0.23536276988912663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.8292,
      "mean_token_accuracy": 0.14440490528941155,
      "num_tokens": 5233829.0,
      "step": 3025
    },
    {
      "entropy": 5.998610305786133,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.9395,
      "mean_token_accuracy": 0.1378275215625763,
      "num_tokens": 5243144.0,
      "step": 3030
    },
    {
      "entropy": 5.9781769752502445,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.7548,
      "mean_token_accuracy": 0.1460421398282051,
      "num_tokens": 5251097.0,
      "step": 3035
    },
    {
      "entropy": 5.962513780593872,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.8404,
      "mean_token_accuracy": 0.15355018228292466,
      "num_tokens": 5260050.0,
      "step": 3040
    },
    {
      "entropy": 5.938783931732178,
      "epoch": 0.2369188873759969,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.645,
      "mean_token_accuracy": 0.15660878717899324,
      "num_tokens": 5268363.0,
      "step": 3045
    },
    {
      "entropy": 5.887162923812866,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1478561833500862,
      "num_tokens": 5276821.0,
      "step": 3050
    },
    {
      "entropy": 5.9527363777160645,
      "epoch": 0.237696946119432,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14752280563116074,
      "num_tokens": 5285419.0,
      "step": 3055
    },
    {
      "entropy": 5.866193866729736,
      "epoch": 0.23808597549114957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.6553,
      "mean_token_accuracy": 0.1510017544031143,
      "num_tokens": 5294004.0,
      "step": 3060
    },
    {
      "entropy": 5.8557802677154545,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.8301,
      "mean_token_accuracy": 0.14936043918132783,
      "num_tokens": 5302699.0,
      "step": 3065
    },
    {
      "entropy": 5.966806888580322,
      "epoch": 0.23886403423458472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499707771473075,
      "loss": 5.7575,
      "mean_token_accuracy": 0.15072435289621353,
      "num_tokens": 5311115.0,
      "step": 3070
    },
    {
      "entropy": 5.901711511611938,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.7648,
      "mean_token_accuracy": 0.15388039052486419,
      "num_tokens": 5319795.0,
      "step": 3075
    },
    {
      "entropy": 5.978531169891357,
      "epoch": 0.23964209297801983,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.8031,
      "mean_token_accuracy": 0.15001416355371475,
      "num_tokens": 5328273.0,
      "step": 3080
    },
    {
      "entropy": 5.9556351661682125,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.125,
      "learning_rate": 0.000499703519804485,
      "loss": 5.7934,
      "mean_token_accuracy": 0.1462474137544632,
      "num_tokens": 5336394.0,
      "step": 3085
    },
    {
      "entropy": 5.87838306427002,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.7355,
      "mean_token_accuracy": 0.15258182883262633,
      "num_tokens": 5344480.0,
      "step": 3090
    },
    {
      "entropy": 5.896993970870971,
      "epoch": 0.24080918109317254,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.797,
      "mean_token_accuracy": 0.141014863550663,
      "num_tokens": 5353445.0,
      "step": 3095
    },
    {
      "entropy": 5.892958879470825,
      "epoch": 0.2411982104648901,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499699237450231,
      "loss": 5.6596,
      "mean_token_accuracy": 0.16075871959328653,
      "num_tokens": 5362356.0,
      "step": 3100
    },
    {
      "entropy": 5.916681575775146,
      "epoch": 0.24158723983660765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.7964,
      "mean_token_accuracy": 0.14848088175058366,
      "num_tokens": 5371467.0,
      "step": 3105
    },
    {
      "entropy": 5.812461185455322,
      "epoch": 0.24197626920832524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.6558,
      "mean_token_accuracy": 0.15510347113013268,
      "num_tokens": 5380256.0,
      "step": 3110
    },
    {
      "entropy": 5.971241188049317,
      "epoch": 0.2423652985800428,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.7135,
      "mean_token_accuracy": 0.15098942071199417,
      "num_tokens": 5388884.0,
      "step": 3115
    },
    {
      "entropy": 5.897061204910278,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499693479912315,
      "loss": 5.6853,
      "mean_token_accuracy": 0.15718052834272384,
      "num_tokens": 5397980.0,
      "step": 3120
    },
    {
      "entropy": 5.859615325927734,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.7922,
      "mean_token_accuracy": 0.14827542155981063,
      "num_tokens": 5406879.0,
      "step": 3125
    },
    {
      "entropy": 5.936743116378784,
      "epoch": 0.24353238669519547,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.7959,
      "mean_token_accuracy": 0.14135870039463044,
      "num_tokens": 5416161.0,
      "step": 3130
    },
    {
      "entropy": 5.887562036514282,
      "epoch": 0.24392141606691306,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.6741,
      "mean_token_accuracy": 0.15898128598928452,
      "num_tokens": 5424323.0,
      "step": 3135
    },
    {
      "entropy": 5.883194923400879,
      "epoch": 0.24431044543863062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.8032,
      "mean_token_accuracy": 0.14068420529365538,
      "num_tokens": 5433750.0,
      "step": 3140
    },
    {
      "entropy": 5.986171770095825,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.821,
      "mean_token_accuracy": 0.14756249487400055,
      "num_tokens": 5442113.0,
      "step": 3145
    },
    {
      "entropy": 5.898403453826904,
      "epoch": 0.24508850418206574,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.6867,
      "mean_token_accuracy": 0.1511594235897064,
      "num_tokens": 5450403.0,
      "step": 3150
    },
    {
      "entropy": 5.896078729629517,
      "epoch": 0.24547753355378332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.7644,
      "mean_token_accuracy": 0.14809978604316712,
      "num_tokens": 5459018.0,
      "step": 3155
    },
    {
      "entropy": 5.9197851657867435,
      "epoch": 0.24586656292550088,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.7983,
      "mean_token_accuracy": 0.14402209520339965,
      "num_tokens": 5468314.0,
      "step": 3160
    },
    {
      "entropy": 5.9504999160766605,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.7793,
      "mean_token_accuracy": 0.1457969695329666,
      "num_tokens": 5476950.0,
      "step": 3165
    },
    {
      "entropy": 5.940018224716186,
      "epoch": 0.246644621668936,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.7464,
      "mean_token_accuracy": 0.1534033253788948,
      "num_tokens": 5485201.0,
      "step": 3170
    },
    {
      "entropy": 6.006494426727295,
      "epoch": 0.24703365104065356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.8628,
      "mean_token_accuracy": 0.14921573102474212,
      "num_tokens": 5493096.0,
      "step": 3175
    },
    {
      "entropy": 5.871295690536499,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.6369,
      "mean_token_accuracy": 0.15377923771739005,
      "num_tokens": 5502413.0,
      "step": 3180
    },
    {
      "entropy": 5.906892967224121,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.781,
      "mean_token_accuracy": 0.15396690219640732,
      "num_tokens": 5511670.0,
      "step": 3185
    },
    {
      "entropy": 5.927646780014038,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.6606,
      "mean_token_accuracy": 0.15735298097133638,
      "num_tokens": 5519344.0,
      "step": 3190
    },
    {
      "entropy": 5.830929899215699,
      "epoch": 0.24858976852752382,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.6786,
      "mean_token_accuracy": 0.14968922585248948,
      "num_tokens": 5528413.0,
      "step": 3195
    },
    {
      "entropy": 5.8730100154876705,
      "epoch": 0.2489787978992414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.7738,
      "mean_token_accuracy": 0.14690972194075586,
      "num_tokens": 5537201.0,
      "step": 3200
    },
    {
      "entropy": 5.923853874206543,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.7804,
      "mean_token_accuracy": 0.14298839941620828,
      "num_tokens": 5545975.0,
      "step": 3205
    },
    {
      "entropy": 5.8805577754974365,
      "epoch": 0.24975685664267652,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.8066,
      "mean_token_accuracy": 0.1442289024591446,
      "num_tokens": 5555237.0,
      "step": 3210
    },
    {
      "entropy": 5.989409494400024,
      "epoch": 0.2501458860143941,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.6911,
      "mean_token_accuracy": 0.15224729627370834,
      "num_tokens": 5564048.0,
      "step": 3215
    },
    {
      "entropy": 5.877629709243775,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.7427,
      "mean_token_accuracy": 0.154438579082489,
      "num_tokens": 5573615.0,
      "step": 3220
    },
    {
      "entropy": 5.876510715484619,
      "epoch": 0.2509239447578292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.6247,
      "mean_token_accuracy": 0.15758016407489778,
      "num_tokens": 5582211.0,
      "step": 3225
    },
    {
      "entropy": 5.906939220428467,
      "epoch": 0.2513129741295468,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.8569,
      "mean_token_accuracy": 0.14443435817956923,
      "num_tokens": 5592542.0,
      "step": 3230
    },
    {
      "entropy": 5.988466691970825,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499659315504938,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15278731137514115,
      "num_tokens": 5601279.0,
      "step": 3235
    },
    {
      "entropy": 5.927760648727417,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1475792944431305,
      "num_tokens": 5610266.0,
      "step": 3240
    },
    {
      "entropy": 5.8933803081512455,
      "epoch": 0.2524800622446995,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.8036,
      "mean_token_accuracy": 0.14620354622602463,
      "num_tokens": 5618799.0,
      "step": 3245
    },
    {
      "entropy": 6.05321683883667,
      "epoch": 0.252869091616417,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.7447,
      "mean_token_accuracy": 0.15469980388879775,
      "num_tokens": 5627861.0,
      "step": 3250
    },
    {
      "entropy": 5.931217432022095,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.7375,
      "mean_token_accuracy": 0.14861182644963264,
      "num_tokens": 5636229.0,
      "step": 3255
    },
    {
      "entropy": 5.938875341415406,
      "epoch": 0.2536471503598522,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499651649830945,
      "loss": 5.9558,
      "mean_token_accuracy": 0.14124528542160988,
      "num_tokens": 5645828.0,
      "step": 3260
    },
    {
      "entropy": 5.8967883586883545,
      "epoch": 0.2540361797315697,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.5671,
      "mean_token_accuracy": 0.15032654106616974,
      "num_tokens": 5654750.0,
      "step": 3265
    },
    {
      "entropy": 5.8799279689788815,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499648559700066,
      "loss": 5.7448,
      "mean_token_accuracy": 0.14809657484292985,
      "num_tokens": 5664017.0,
      "step": 3270
    },
    {
      "entropy": 5.874881553649902,
      "epoch": 0.25481423847500484,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.7406,
      "mean_token_accuracy": 0.14475095346570016,
      "num_tokens": 5672220.0,
      "step": 3275
    },
    {
      "entropy": 5.830359411239624,
      "epoch": 0.2552032678467224,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.6864,
      "mean_token_accuracy": 0.1500364437699318,
      "num_tokens": 5680921.0,
      "step": 3280
    },
    {
      "entropy": 5.955457019805908,
      "epoch": 0.25559229721844,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1482978232204914,
      "num_tokens": 5689226.0,
      "step": 3285
    },
    {
      "entropy": 5.929180049896241,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.8345,
      "mean_token_accuracy": 0.143748040497303,
      "num_tokens": 5698413.0,
      "step": 3290
    },
    {
      "entropy": 5.851541137695312,
      "epoch": 0.25637035596187513,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.718,
      "mean_token_accuracy": 0.14506666213274003,
      "num_tokens": 5707724.0,
      "step": 3295
    },
    {
      "entropy": 5.939691019058228,
      "epoch": 0.25675938533359266,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049963920749976,
      "loss": 5.7271,
      "mean_token_accuracy": 0.14950018227100373,
      "num_tokens": 5716566.0,
      "step": 3300
    },
    {
      "entropy": 5.84193868637085,
      "epoch": 0.25714841470531025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.6666,
      "mean_token_accuracy": 0.1517062321305275,
      "num_tokens": 5725463.0,
      "step": 3305
    },
    {
      "entropy": 5.895420980453491,
      "epoch": 0.25753744407702783,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.6642,
      "mean_token_accuracy": 0.1558502957224846,
      "num_tokens": 5734143.0,
      "step": 3310
    },
    {
      "entropy": 5.89042387008667,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.7478,
      "mean_token_accuracy": 0.15216899216175078,
      "num_tokens": 5742981.0,
      "step": 3315
    },
    {
      "entropy": 5.7684619426727295,
      "epoch": 0.25831550282046295,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.625,
      "mean_token_accuracy": 0.1628153458237648,
      "num_tokens": 5751294.0,
      "step": 3320
    },
    {
      "entropy": 5.869237899780273,
      "epoch": 0.25870453219218054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.7406,
      "mean_token_accuracy": 0.1526793658733368,
      "num_tokens": 5760958.0,
      "step": 3325
    },
    {
      "entropy": 5.935721826553345,
      "epoch": 0.25909356156389807,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.7595,
      "mean_token_accuracy": 0.1518444985151291,
      "num_tokens": 5769077.0,
      "step": 3330
    },
    {
      "entropy": 5.96410346031189,
      "epoch": 0.25948259093561565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.7934,
      "mean_token_accuracy": 0.14791057929396628,
      "num_tokens": 5778093.0,
      "step": 3335
    },
    {
      "entropy": 5.904092693328858,
      "epoch": 0.2598716203073332,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.7365,
      "mean_token_accuracy": 0.15056012198328972,
      "num_tokens": 5786996.0,
      "step": 3340
    },
    {
      "entropy": 5.893521738052368,
      "epoch": 0.26026064967905077,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.7462,
      "mean_token_accuracy": 0.1535051003098488,
      "num_tokens": 5794831.0,
      "step": 3345
    },
    {
      "entropy": 5.789423561096191,
      "epoch": 0.26064967905076836,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.5761,
      "mean_token_accuracy": 0.15884069055318834,
      "num_tokens": 5802286.0,
      "step": 3350
    },
    {
      "entropy": 5.788263034820557,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.7195,
      "mean_token_accuracy": 0.15141943246126174,
      "num_tokens": 5810985.0,
      "step": 3355
    },
    {
      "entropy": 5.955795001983643,
      "epoch": 0.2614277377942035,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499620134981758,
      "loss": 5.7707,
      "mean_token_accuracy": 0.15107223987579346,
      "num_tokens": 5820037.0,
      "step": 3360
    },
    {
      "entropy": 5.795082426071167,
      "epoch": 0.261816767165921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.6242,
      "mean_token_accuracy": 0.16446445733308793,
      "num_tokens": 5828232.0,
      "step": 3365
    },
    {
      "entropy": 5.909881448745727,
      "epoch": 0.2622057965376386,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15064912438392639,
      "num_tokens": 5838184.0,
      "step": 3370
    },
    {
      "entropy": 5.951297903060913,
      "epoch": 0.2625948259093562,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.6708,
      "mean_token_accuracy": 0.15528088957071304,
      "num_tokens": 5846705.0,
      "step": 3375
    },
    {
      "entropy": 5.891704511642456,
      "epoch": 0.2629838552810737,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.7557,
      "mean_token_accuracy": 0.14960285723209382,
      "num_tokens": 5855959.0,
      "step": 3380
    },
    {
      "entropy": 5.8450446128845215,
      "epoch": 0.2633728846527913,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.7099,
      "mean_token_accuracy": 0.15203075855970383,
      "num_tokens": 5865283.0,
      "step": 3385
    },
    {
      "entropy": 5.913759851455689,
      "epoch": 0.2637619140245088,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.7391,
      "mean_token_accuracy": 0.1565643608570099,
      "num_tokens": 5873833.0,
      "step": 3390
    },
    {
      "entropy": 5.948887634277344,
      "epoch": 0.2641509433962264,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.6684,
      "mean_token_accuracy": 0.16087514907121658,
      "num_tokens": 5882027.0,
      "step": 3395
    },
    {
      "entropy": 5.710987138748169,
      "epoch": 0.264539972767944,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.548,
      "mean_token_accuracy": 0.16150110214948654,
      "num_tokens": 5890650.0,
      "step": 3400
    },
    {
      "entropy": 5.778266859054566,
      "epoch": 0.26492900213966153,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.6004,
      "mean_token_accuracy": 0.16101501286029815,
      "num_tokens": 5898595.0,
      "step": 3405
    },
    {
      "entropy": 5.916549825668335,
      "epoch": 0.2653180315113791,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.7157,
      "mean_token_accuracy": 0.1523623764514923,
      "num_tokens": 5907234.0,
      "step": 3410
    },
    {
      "entropy": 5.877885484695435,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.6562,
      "mean_token_accuracy": 0.15936153680086135,
      "num_tokens": 5915690.0,
      "step": 3415
    },
    {
      "entropy": 5.798541498184204,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.6053,
      "mean_token_accuracy": 0.16079822927713394,
      "num_tokens": 5924481.0,
      "step": 3420
    },
    {
      "entropy": 5.808663415908813,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499598919247571,
      "loss": 5.6951,
      "mean_token_accuracy": 0.1579988867044449,
      "num_tokens": 5933059.0,
      "step": 3425
    },
    {
      "entropy": 5.887850189208985,
      "epoch": 0.26687414899824935,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.8016,
      "mean_token_accuracy": 0.14508845061063766,
      "num_tokens": 5941767.0,
      "step": 3430
    },
    {
      "entropy": 5.844564151763916,
      "epoch": 0.26726317836996694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.691,
      "mean_token_accuracy": 0.15415164232254028,
      "num_tokens": 5950202.0,
      "step": 3435
    },
    {
      "entropy": 5.848789262771606,
      "epoch": 0.2676522077416845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.7359,
      "mean_token_accuracy": 0.14812117516994477,
      "num_tokens": 5959242.0,
      "step": 3440
    },
    {
      "entropy": 5.846918964385987,
      "epoch": 0.26804123711340205,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.6836,
      "mean_token_accuracy": 0.15515549182891847,
      "num_tokens": 5968404.0,
      "step": 3445
    },
    {
      "entropy": 5.860494327545166,
      "epoch": 0.26843026648511964,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.8005,
      "mean_token_accuracy": 0.15509908199310302,
      "num_tokens": 5976711.0,
      "step": 3450
    },
    {
      "entropy": 5.894256687164306,
      "epoch": 0.26881929585683717,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.6551,
      "mean_token_accuracy": 0.14889176189899445,
      "num_tokens": 5985049.0,
      "step": 3455
    },
    {
      "entropy": 5.8000787734985355,
      "epoch": 0.26920832522855476,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.6823,
      "mean_token_accuracy": 0.15802373141050338,
      "num_tokens": 5993904.0,
      "step": 3460
    },
    {
      "entropy": 5.915686511993409,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.7404,
      "mean_token_accuracy": 0.15314059257507323,
      "num_tokens": 6001832.0,
      "step": 3465
    },
    {
      "entropy": 5.825476264953613,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1504707530140877,
      "num_tokens": 6010609.0,
      "step": 3470
    },
    {
      "entropy": 5.851345634460449,
      "epoch": 0.27037541334370746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.7193,
      "mean_token_accuracy": 0.15108724534511567,
      "num_tokens": 6019299.0,
      "step": 3475
    },
    {
      "entropy": 5.943111848831177,
      "epoch": 0.270764442715425,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.7676,
      "mean_token_accuracy": 0.14575971737504007,
      "num_tokens": 6028743.0,
      "step": 3480
    },
    {
      "entropy": 5.879656744003296,
      "epoch": 0.2711534720871426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.6713,
      "mean_token_accuracy": 0.14724702388048172,
      "num_tokens": 6038064.0,
      "step": 3485
    },
    {
      "entropy": 5.793353033065796,
      "epoch": 0.27154250145886016,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.6612,
      "mean_token_accuracy": 0.1524479627609253,
      "num_tokens": 6047031.0,
      "step": 3490
    },
    {
      "entropy": 5.889426136016846,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.7498,
      "mean_token_accuracy": 0.15119700282812118,
      "num_tokens": 6056147.0,
      "step": 3495
    },
    {
      "entropy": 5.890491199493408,
      "epoch": 0.2723205602022953,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.6689,
      "mean_token_accuracy": 0.15356390327215194,
      "num_tokens": 6064641.0,
      "step": 3500
    },
    {
      "entropy": 5.796261215209961,
      "epoch": 0.2727095895740128,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15325731337070464,
      "num_tokens": 6073664.0,
      "step": 3505
    },
    {
      "entropy": 5.811582803726196,
      "epoch": 0.2730986189457304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.5752,
      "mean_token_accuracy": 0.16186247318983077,
      "num_tokens": 6081480.0,
      "step": 3510
    },
    {
      "entropy": 5.8565483570098875,
      "epoch": 0.273487648317448,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.7021,
      "mean_token_accuracy": 0.14859129190444947,
      "num_tokens": 6090408.0,
      "step": 3515
    },
    {
      "entropy": 5.866174983978271,
      "epoch": 0.2738766776891655,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499566875640638,
      "loss": 5.699,
      "mean_token_accuracy": 0.15143164694309236,
      "num_tokens": 6099076.0,
      "step": 3520
    },
    {
      "entropy": 5.874391031265259,
      "epoch": 0.2742657070608831,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.7466,
      "mean_token_accuracy": 0.14636368602514266,
      "num_tokens": 6108346.0,
      "step": 3525
    },
    {
      "entropy": 5.837771892547607,
      "epoch": 0.2746547364326007,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.7037,
      "mean_token_accuracy": 0.15670324489474297,
      "num_tokens": 6116974.0,
      "step": 3530
    },
    {
      "entropy": 5.913712024688721,
      "epoch": 0.2750437658043182,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.7323,
      "mean_token_accuracy": 0.1500618875026703,
      "num_tokens": 6125510.0,
      "step": 3535
    },
    {
      "entropy": 5.865264272689819,
      "epoch": 0.2754327951760358,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.7712,
      "mean_token_accuracy": 0.14675354063510895,
      "num_tokens": 6134959.0,
      "step": 3540
    },
    {
      "entropy": 5.930070304870606,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.7049,
      "mean_token_accuracy": 0.145357196778059,
      "num_tokens": 6144485.0,
      "step": 3545
    },
    {
      "entropy": 5.86058931350708,
      "epoch": 0.2762108539194709,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.6487,
      "mean_token_accuracy": 0.15821592509746552,
      "num_tokens": 6153991.0,
      "step": 3550
    },
    {
      "entropy": 5.739012575149536,
      "epoch": 0.2765998832911885,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.6425,
      "mean_token_accuracy": 0.1544785276055336,
      "num_tokens": 6162649.0,
      "step": 3555
    },
    {
      "entropy": 5.813714075088501,
      "epoch": 0.27698891266290604,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.5463,
      "mean_token_accuracy": 0.15806780904531478,
      "num_tokens": 6171525.0,
      "step": 3560
    },
    {
      "entropy": 5.872299385070801,
      "epoch": 0.2773779420346236,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.7726,
      "mean_token_accuracy": 0.15298861116170884,
      "num_tokens": 6180403.0,
      "step": 3565
    },
    {
      "entropy": 5.834949064254761,
      "epoch": 0.27776697140634116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.6142,
      "mean_token_accuracy": 0.15946034938097,
      "num_tokens": 6188361.0,
      "step": 3570
    },
    {
      "entropy": 5.807054090499878,
      "epoch": 0.27815600077805874,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.6716,
      "mean_token_accuracy": 0.14874202236533166,
      "num_tokens": 6197418.0,
      "step": 3575
    },
    {
      "entropy": 5.810306215286255,
      "epoch": 0.27854503014977633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.6813,
      "mean_token_accuracy": 0.15601591318845748,
      "num_tokens": 6206040.0,
      "step": 3580
    },
    {
      "entropy": 5.79304084777832,
      "epoch": 0.27893405952149386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.5968,
      "mean_token_accuracy": 0.15546937584877013,
      "num_tokens": 6214433.0,
      "step": 3585
    },
    {
      "entropy": 5.769221782684326,
      "epoch": 0.27932308889321145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.5765,
      "mean_token_accuracy": 0.15947029739618301,
      "num_tokens": 6223375.0,
      "step": 3590
    },
    {
      "entropy": 5.744332695007325,
      "epoch": 0.279712118264929,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499540709149229,
      "loss": 5.6898,
      "mean_token_accuracy": 0.15379903316497803,
      "num_tokens": 6232597.0,
      "step": 3595
    },
    {
      "entropy": 5.785550165176391,
      "epoch": 0.28010114763664656,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.5429,
      "mean_token_accuracy": 0.16784969419240953,
      "num_tokens": 6240634.0,
      "step": 3600
    },
    {
      "entropy": 5.780906867980957,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.5015,
      "mean_token_accuracy": 0.15682514160871505,
      "num_tokens": 6249098.0,
      "step": 3605
    },
    {
      "entropy": 5.739667129516602,
      "epoch": 0.2808792063800817,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499535383857776,
      "loss": 5.5472,
      "mean_token_accuracy": 0.16006743013858796,
      "num_tokens": 6257572.0,
      "step": 3610
    },
    {
      "entropy": 5.859043931961059,
      "epoch": 0.28126823575179927,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499533601946655,
      "loss": 5.8165,
      "mean_token_accuracy": 0.14751632660627365,
      "num_tokens": 6266492.0,
      "step": 3615
    },
    {
      "entropy": 5.865448904037476,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.6829,
      "mean_token_accuracy": 0.155305153131485,
      "num_tokens": 6275149.0,
      "step": 3620
    },
    {
      "entropy": 5.874626731872558,
      "epoch": 0.2820462944952344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.6956,
      "mean_token_accuracy": 0.15312935113906861,
      "num_tokens": 6283208.0,
      "step": 3625
    },
    {
      "entropy": 5.855358982086182,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049952823577176,
      "loss": 5.6347,
      "mean_token_accuracy": 0.15589332580566406,
      "num_tokens": 6292084.0,
      "step": 3630
    },
    {
      "entropy": 5.784860134124756,
      "epoch": 0.2828243532386695,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.6336,
      "mean_token_accuracy": 0.1528566986322403,
      "num_tokens": 6301855.0,
      "step": 3635
    },
    {
      "entropy": 5.845756483078003,
      "epoch": 0.2832133826103871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.6379,
      "mean_token_accuracy": 0.15769062787294388,
      "num_tokens": 6310166.0,
      "step": 3640
    },
    {
      "entropy": 5.783097553253174,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499522838935138,
      "loss": 5.6228,
      "mean_token_accuracy": 0.16248167604207991,
      "num_tokens": 6318928.0,
      "step": 3645
    },
    {
      "entropy": 5.826745843887329,
      "epoch": 0.2839914413538222,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.6509,
      "mean_token_accuracy": 0.1659294068813324,
      "num_tokens": 6327790.0,
      "step": 3650
    },
    {
      "entropy": 5.85683012008667,
      "epoch": 0.2843804707255398,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1522120863199234,
      "num_tokens": 6335517.0,
      "step": 3655
    },
    {
      "entropy": 5.789180564880371,
      "epoch": 0.2847695000972573,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.7049,
      "mean_token_accuracy": 0.15419832319021226,
      "num_tokens": 6344218.0,
      "step": 3660
    },
    {
      "entropy": 5.769765949249267,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.5136,
      "mean_token_accuracy": 0.16411114037036895,
      "num_tokens": 6353125.0,
      "step": 3665
    },
    {
      "entropy": 5.8027407169342045,
      "epoch": 0.2855475588406925,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.6528,
      "mean_token_accuracy": 0.1551785096526146,
      "num_tokens": 6362113.0,
      "step": 3670
    },
    {
      "entropy": 5.844509744644165,
      "epoch": 0.28593658821241,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.7323,
      "mean_token_accuracy": 0.15433150082826613,
      "num_tokens": 6370916.0,
      "step": 3675
    },
    {
      "entropy": 5.866036510467529,
      "epoch": 0.2863256175841276,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499510127080451,
      "loss": 5.5915,
      "mean_token_accuracy": 0.1586878553032875,
      "num_tokens": 6379256.0,
      "step": 3680
    },
    {
      "entropy": 5.825314474105835,
      "epoch": 0.28671464695584514,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.723,
      "mean_token_accuracy": 0.15312020778656005,
      "num_tokens": 6388098.0,
      "step": 3685
    },
    {
      "entropy": 5.7031796932220455,
      "epoch": 0.28710367632756273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.5249,
      "mean_token_accuracy": 0.16168365329504014,
      "num_tokens": 6396292.0,
      "step": 3690
    },
    {
      "entropy": 5.792140960693359,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.5502,
      "mean_token_accuracy": 0.16206852197647095,
      "num_tokens": 6404968.0,
      "step": 3695
    },
    {
      "entropy": 5.691653776168823,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.6329,
      "mean_token_accuracy": 0.15900261849164962,
      "num_tokens": 6412839.0,
      "step": 3700
    },
    {
      "entropy": 5.784019470214844,
      "epoch": 0.28827076444271543,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.6493,
      "mean_token_accuracy": 0.14958156421780586,
      "num_tokens": 6421943.0,
      "step": 3705
    },
    {
      "entropy": 5.855062103271484,
      "epoch": 0.28865979381443296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499499098347882,
      "loss": 5.7285,
      "mean_token_accuracy": 0.14923914447426795,
      "num_tokens": 6430731.0,
      "step": 3710
    },
    {
      "entropy": 5.851906061172485,
      "epoch": 0.28904882318615055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.6556,
      "mean_token_accuracy": 0.1556018888950348,
      "num_tokens": 6439365.0,
      "step": 3715
    },
    {
      "entropy": 5.8089118003845215,
      "epoch": 0.28943785255786814,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499495394852062,
      "loss": 5.6021,
      "mean_token_accuracy": 0.16112566590309144,
      "num_tokens": 6448085.0,
      "step": 3720
    },
    {
      "entropy": 5.684608078002929,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.6652,
      "mean_token_accuracy": 0.14997439682483674,
      "num_tokens": 6457630.0,
      "step": 3725
    },
    {
      "entropy": 5.838469409942627,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1629138931632042,
      "num_tokens": 6465781.0,
      "step": 3730
    },
    {
      "entropy": 5.844767808914185,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.7055,
      "mean_token_accuracy": 0.1529698833823204,
      "num_tokens": 6473723.0,
      "step": 3735
    },
    {
      "entropy": 5.711420583724975,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.6375,
      "mean_token_accuracy": 0.1572704315185547,
      "num_tokens": 6482535.0,
      "step": 3740
    },
    {
      "entropy": 5.8504077911376955,
      "epoch": 0.29138299941645596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.6747,
      "mean_token_accuracy": 0.1574747622013092,
      "num_tokens": 6491297.0,
      "step": 3745
    },
    {
      "entropy": 5.773846578598023,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499484202612859,
      "loss": 5.6295,
      "mean_token_accuracy": 0.15734697729349137,
      "num_tokens": 6499809.0,
      "step": 3750
    },
    {
      "entropy": 5.7556627750396725,
      "epoch": 0.2921610581598911,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.6075,
      "mean_token_accuracy": 0.16184348315000535,
      "num_tokens": 6507965.0,
      "step": 3755
    },
    {
      "entropy": 5.852641582489014,
      "epoch": 0.29255008753160866,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.6552,
      "mean_token_accuracy": 0.15120026841759682,
      "num_tokens": 6517106.0,
      "step": 3760
    },
    {
      "entropy": 5.845433378219605,
      "epoch": 0.2929391169033262,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.7337,
      "mean_token_accuracy": 0.15373013466596602,
      "num_tokens": 6526663.0,
      "step": 3765
    },
    {
      "entropy": 5.927438497543335,
      "epoch": 0.2933281462750438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1520956665277481,
      "num_tokens": 6535732.0,
      "step": 3770
    },
    {
      "entropy": 5.808690547943115,
      "epoch": 0.2937171756467613,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.6301,
      "mean_token_accuracy": 0.15487213283777237,
      "num_tokens": 6544938.0,
      "step": 3775
    },
    {
      "entropy": 5.802787160873413,
      "epoch": 0.2941062050184789,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.6336,
      "mean_token_accuracy": 0.1528394863009453,
      "num_tokens": 6554507.0,
      "step": 3780
    },
    {
      "entropy": 5.839257526397705,
      "epoch": 0.2944952343901965,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.6781,
      "mean_token_accuracy": 0.15294660478830338,
      "num_tokens": 6563285.0,
      "step": 3785
    },
    {
      "entropy": 5.752003622055054,
      "epoch": 0.294884263761914,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.56,
      "mean_token_accuracy": 0.15720225274562835,
      "num_tokens": 6572223.0,
      "step": 3790
    },
    {
      "entropy": 5.823800420761108,
      "epoch": 0.2952732931336316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.7438,
      "mean_token_accuracy": 0.14936941713094712,
      "num_tokens": 6580686.0,
      "step": 3795
    },
    {
      "entropy": 5.855613756179809,
      "epoch": 0.29566232250534913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.6053,
      "mean_token_accuracy": 0.1560147821903229,
      "num_tokens": 6589056.0,
      "step": 3800
    },
    {
      "entropy": 5.8202354431152346,
      "epoch": 0.2960513518770667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.6695,
      "mean_token_accuracy": 0.15734007209539413,
      "num_tokens": 6598221.0,
      "step": 3805
    },
    {
      "entropy": 5.879884719848633,
      "epoch": 0.2964403812487843,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.72,
      "mean_token_accuracy": 0.15723774582147598,
      "num_tokens": 6606657.0,
      "step": 3810
    },
    {
      "entropy": 5.883637285232544,
      "epoch": 0.29682941062050183,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.8057,
      "mean_token_accuracy": 0.14732917174696922,
      "num_tokens": 6616203.0,
      "step": 3815
    },
    {
      "entropy": 5.904540967941284,
      "epoch": 0.2972184399922194,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.6617,
      "mean_token_accuracy": 0.15442878901958465,
      "num_tokens": 6624982.0,
      "step": 3820
    },
    {
      "entropy": 5.843383693695069,
      "epoch": 0.29760746936393695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.6234,
      "mean_token_accuracy": 0.16022102981805803,
      "num_tokens": 6632548.0,
      "step": 3825
    },
    {
      "entropy": 5.7950318336486815,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.7647,
      "mean_token_accuracy": 0.14879212528467178,
      "num_tokens": 6641943.0,
      "step": 3830
    },
    {
      "entropy": 5.838181400299073,
      "epoch": 0.2983855281073721,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.6,
      "mean_token_accuracy": 0.16140060871839523,
      "num_tokens": 6650371.0,
      "step": 3835
    },
    {
      "entropy": 5.819985818862915,
      "epoch": 0.29877455747908965,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.6843,
      "mean_token_accuracy": 0.15791809260845185,
      "num_tokens": 6659324.0,
      "step": 3840
    },
    {
      "entropy": 5.810818195343018,
      "epoch": 0.29916358685080724,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.6492,
      "mean_token_accuracy": 0.1515626162290573,
      "num_tokens": 6668237.0,
      "step": 3845
    },
    {
      "entropy": 5.808945417404175,
      "epoch": 0.2995526162225248,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.5772,
      "mean_token_accuracy": 0.16233526021242142,
      "num_tokens": 6677213.0,
      "step": 3850
    },
    {
      "entropy": 5.7874250411987305,
      "epoch": 0.29994164559424236,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499444064161418,
      "loss": 5.587,
      "mean_token_accuracy": 0.15759442001581192,
      "num_tokens": 6686850.0,
      "step": 3855
    },
    {
      "entropy": 5.77463321685791,
      "epoch": 0.30033067496595994,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499442115343045,
      "loss": 5.5608,
      "mean_token_accuracy": 0.16466309428215026,
      "num_tokens": 6695697.0,
      "step": 3860
    },
    {
      "entropy": 5.764589834213257,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.6185,
      "mean_token_accuracy": 0.16456811875104904,
      "num_tokens": 6704525.0,
      "step": 3865
    },
    {
      "entropy": 5.749872732162475,
      "epoch": 0.30110873370939506,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.555,
      "mean_token_accuracy": 0.1628976732492447,
      "num_tokens": 6713562.0,
      "step": 3870
    },
    {
      "entropy": 5.841018533706665,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.6633,
      "mean_token_accuracy": 0.15924778133630751,
      "num_tokens": 6721817.0,
      "step": 3875
    },
    {
      "entropy": 5.8318479537963865,
      "epoch": 0.3018867924528302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.5799,
      "mean_token_accuracy": 0.16334821432828903,
      "num_tokens": 6729757.0,
      "step": 3880
    },
    {
      "entropy": 5.756866598129273,
      "epoch": 0.30227582182454776,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.6585,
      "mean_token_accuracy": 0.1475125730037689,
      "num_tokens": 6738354.0,
      "step": 3885
    },
    {
      "entropy": 5.819000101089477,
      "epoch": 0.3026648511962653,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.7076,
      "mean_token_accuracy": 0.15129155218601226,
      "num_tokens": 6746505.0,
      "step": 3890
    },
    {
      "entropy": 5.904794931411743,
      "epoch": 0.3030538805679829,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.6111,
      "mean_token_accuracy": 0.15827509462833406,
      "num_tokens": 6754547.0,
      "step": 3895
    },
    {
      "entropy": 5.7554058074951175,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.5293,
      "mean_token_accuracy": 0.16018292903900147,
      "num_tokens": 6763213.0,
      "step": 3900
    },
    {
      "entropy": 5.691382646560669,
      "epoch": 0.303831939311418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.5787,
      "mean_token_accuracy": 0.15872986018657684,
      "num_tokens": 6771060.0,
      "step": 3905
    },
    {
      "entropy": 5.808207893371582,
      "epoch": 0.3042209686831356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.7216,
      "mean_token_accuracy": 0.15398526936769485,
      "num_tokens": 6779819.0,
      "step": 3910
    },
    {
      "entropy": 5.835114002227783,
      "epoch": 0.3046099980548531,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.618,
      "mean_token_accuracy": 0.15699677169322968,
      "num_tokens": 6788660.0,
      "step": 3915
    },
    {
      "entropy": 5.723716831207275,
      "epoch": 0.3049990274265707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.5708,
      "mean_token_accuracy": 0.16314460784196855,
      "num_tokens": 6797298.0,
      "step": 3920
    },
    {
      "entropy": 5.743786954879761,
      "epoch": 0.3053880567982883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.6223,
      "mean_token_accuracy": 0.15216543674468994,
      "num_tokens": 6806148.0,
      "step": 3925
    },
    {
      "entropy": 5.764825057983399,
      "epoch": 0.3057770861700058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499414474316426,
      "loss": 5.6592,
      "mean_token_accuracy": 0.157614828646183,
      "num_tokens": 6815494.0,
      "step": 3930
    },
    {
      "entropy": 5.904352045059204,
      "epoch": 0.3061661155417234,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.7051,
      "mean_token_accuracy": 0.15148080736398697,
      "num_tokens": 6824852.0,
      "step": 3935
    },
    {
      "entropy": 5.787218236923218,
      "epoch": 0.306555144913441,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16816734969615937,
      "num_tokens": 6834200.0,
      "step": 3940
    },
    {
      "entropy": 5.774364185333252,
      "epoch": 0.3069441742851585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.6567,
      "mean_token_accuracy": 0.15507219433784486,
      "num_tokens": 6842869.0,
      "step": 3945
    },
    {
      "entropy": 5.814112663269043,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1599549800157547,
      "num_tokens": 6851668.0,
      "step": 3950
    },
    {
      "entropy": 5.831446838378906,
      "epoch": 0.30772223302859364,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.6875,
      "mean_token_accuracy": 0.16353656947612763,
      "num_tokens": 6860710.0,
      "step": 3955
    },
    {
      "entropy": 5.75144510269165,
      "epoch": 0.3081112624003112,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.6216,
      "mean_token_accuracy": 0.1537766396999359,
      "num_tokens": 6869608.0,
      "step": 3960
    },
    {
      "entropy": 5.827413606643677,
      "epoch": 0.3085002917720288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.639,
      "mean_token_accuracy": 0.1554042413830757,
      "num_tokens": 6878108.0,
      "step": 3965
    },
    {
      "entropy": 5.779709196090698,
      "epoch": 0.30888932114374634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.5448,
      "mean_token_accuracy": 0.15877590030431749,
      "num_tokens": 6886369.0,
      "step": 3970
    },
    {
      "entropy": 5.7670244693756105,
      "epoch": 0.30927835051546393,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.607,
      "mean_token_accuracy": 0.15788756608963012,
      "num_tokens": 6895410.0,
      "step": 3975
    },
    {
      "entropy": 5.797284984588623,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499394322105318,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15986649543046952,
      "num_tokens": 6904380.0,
      "step": 3980
    },
    {
      "entropy": 5.797864580154419,
      "epoch": 0.31005640925889905,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.6206,
      "mean_token_accuracy": 0.15683696866035463,
      "num_tokens": 6913012.0,
      "step": 3985
    },
    {
      "entropy": 5.7702795505523685,
      "epoch": 0.31044543863061663,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.6387,
      "mean_token_accuracy": 0.15548964440822602,
      "num_tokens": 6921696.0,
      "step": 3990
    },
    {
      "entropy": 5.791179227828979,
      "epoch": 0.31083446800233416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499388210046469,
      "loss": 5.579,
      "mean_token_accuracy": 0.15531976968050004,
      "num_tokens": 6930017.0,
      "step": 3995
    },
    {
      "entropy": 5.709838199615478,
      "epoch": 0.31122349737405175,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1598093330860138,
      "num_tokens": 6938410.0,
      "step": 4000
    },
    {
      "entropy": 5.782442045211792,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.6157,
      "mean_token_accuracy": 0.15706539303064346,
      "num_tokens": 6946227.0,
      "step": 4005
    },
    {
      "entropy": 5.797301340103149,
      "epoch": 0.31200155611748687,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.6096,
      "mean_token_accuracy": 0.15807232558727263,
      "num_tokens": 6954747.0,
      "step": 4010
    },
    {
      "entropy": 5.788407516479492,
      "epoch": 0.31239058548920445,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.6804,
      "mean_token_accuracy": 0.15833715796470643,
      "num_tokens": 6962990.0,
      "step": 4015
    },
    {
      "entropy": 5.777428436279297,
      "epoch": 0.312779614860922,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.638,
      "mean_token_accuracy": 0.1595762073993683,
      "num_tokens": 6972351.0,
      "step": 4020
    },
    {
      "entropy": 5.789421892166137,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.5714,
      "mean_token_accuracy": 0.16144972741603852,
      "num_tokens": 6980387.0,
      "step": 4025
    },
    {
      "entropy": 5.658299160003662,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499373829411585,
      "loss": 5.6357,
      "mean_token_accuracy": 0.1596735969185829,
      "num_tokens": 6989388.0,
      "step": 4030
    },
    {
      "entropy": 5.87131175994873,
      "epoch": 0.3139467029760747,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049937176141686,
      "loss": 5.5914,
      "mean_token_accuracy": 0.16546436548233032,
      "num_tokens": 6997714.0,
      "step": 4035
    },
    {
      "entropy": 5.819481420516968,
      "epoch": 0.3143357323477923,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.5748,
      "mean_token_accuracy": 0.16345698684453963,
      "num_tokens": 7006453.0,
      "step": 4040
    },
    {
      "entropy": 5.764643812179566,
      "epoch": 0.3147247617195098,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.6514,
      "mean_token_accuracy": 0.15526093393564225,
      "num_tokens": 7015398.0,
      "step": 4045
    },
    {
      "entropy": 5.767642402648926,
      "epoch": 0.3151137910912274,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.5952,
      "mean_token_accuracy": 0.1578880578279495,
      "num_tokens": 7023796.0,
      "step": 4050
    },
    {
      "entropy": 5.703687334060669,
      "epoch": 0.315502820462945,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.5446,
      "mean_token_accuracy": 0.1635144382715225,
      "num_tokens": 7032453.0,
      "step": 4055
    },
    {
      "entropy": 5.7386377334594725,
      "epoch": 0.3158918498346625,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.6841,
      "mean_token_accuracy": 0.15243002027273178,
      "num_tokens": 7041034.0,
      "step": 4060
    },
    {
      "entropy": 5.8850563049316404,
      "epoch": 0.3162808792063801,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.7752,
      "mean_token_accuracy": 0.1502375230193138,
      "num_tokens": 7050289.0,
      "step": 4065
    },
    {
      "entropy": 5.781525039672852,
      "epoch": 0.3166699085780976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.6123,
      "mean_token_accuracy": 0.16497506499290465,
      "num_tokens": 7059638.0,
      "step": 4070
    },
    {
      "entropy": 5.751022434234619,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.622,
      "mean_token_accuracy": 0.15959509760141372,
      "num_tokens": 7067763.0,
      "step": 4075
    },
    {
      "entropy": 5.7526812076568605,
      "epoch": 0.3174479673215328,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.5854,
      "mean_token_accuracy": 0.16497846245765685,
      "num_tokens": 7076126.0,
      "step": 4080
    },
    {
      "entropy": 5.754148960113525,
      "epoch": 0.31783699669325033,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.5271,
      "mean_token_accuracy": 0.16019581109285355,
      "num_tokens": 7084403.0,
      "step": 4085
    },
    {
      "entropy": 5.6758214950561525,
      "epoch": 0.3182260260649679,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.4989,
      "mean_token_accuracy": 0.1651519536972046,
      "num_tokens": 7093450.0,
      "step": 4090
    },
    {
      "entropy": 5.695490217208862,
      "epoch": 0.31861505543668545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499346679939406,
      "loss": 5.4938,
      "mean_token_accuracy": 0.1639673426747322,
      "num_tokens": 7101722.0,
      "step": 4095
    },
    {
      "entropy": 5.70269193649292,
      "epoch": 0.31900408480840303,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.5973,
      "mean_token_accuracy": 0.15606338530778885,
      "num_tokens": 7111285.0,
      "step": 4100
    },
    {
      "entropy": 5.855121707916259,
      "epoch": 0.3193931141801206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.6664,
      "mean_token_accuracy": 0.1561015322804451,
      "num_tokens": 7120281.0,
      "step": 4105
    },
    {
      "entropy": 5.717962598800659,
      "epoch": 0.31978214355183815,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.5438,
      "mean_token_accuracy": 0.16015197783708573,
      "num_tokens": 7128584.0,
      "step": 4110
    },
    {
      "entropy": 5.808613395690918,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.6513,
      "mean_token_accuracy": 0.15905333682894707,
      "num_tokens": 7137570.0,
      "step": 4115
    },
    {
      "entropy": 5.773133039474487,
      "epoch": 0.32056020229527327,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.6899,
      "mean_token_accuracy": 0.15812519043684006,
      "num_tokens": 7146965.0,
      "step": 4120
    },
    {
      "entropy": 5.787432909011841,
      "epoch": 0.32094923166699085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.6272,
      "mean_token_accuracy": 0.15931697189807892,
      "num_tokens": 7155657.0,
      "step": 4125
    },
    {
      "entropy": 5.8046619415283205,
      "epoch": 0.32133826103870844,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1596067264676094,
      "num_tokens": 7164369.0,
      "step": 4130
    },
    {
      "entropy": 5.7777173042297365,
      "epoch": 0.32172729041042597,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1697876438498497,
      "num_tokens": 7172161.0,
      "step": 4135
    },
    {
      "entropy": 5.706020784378052,
      "epoch": 0.32211631978214356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15171700865030288,
      "num_tokens": 7180606.0,
      "step": 4140
    },
    {
      "entropy": 5.728547096252441,
      "epoch": 0.32250534915386114,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1590207114815712,
      "num_tokens": 7188912.0,
      "step": 4145
    },
    {
      "entropy": 5.788024091720581,
      "epoch": 0.3228943785255787,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499323257964666,
      "loss": 5.6215,
      "mean_token_accuracy": 0.15666755810379981,
      "num_tokens": 7197477.0,
      "step": 4150
    },
    {
      "entropy": 5.801720523834229,
      "epoch": 0.32328340789729626,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.6026,
      "mean_token_accuracy": 0.16264097541570663,
      "num_tokens": 7206571.0,
      "step": 4155
    },
    {
      "entropy": 5.751149368286133,
      "epoch": 0.3236724372690138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.4462,
      "mean_token_accuracy": 0.15989296287298202,
      "num_tokens": 7215319.0,
      "step": 4160
    },
    {
      "entropy": 5.73461594581604,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.6286,
      "mean_token_accuracy": 0.1550238996744156,
      "num_tokens": 7223389.0,
      "step": 4165
    },
    {
      "entropy": 5.771875619888306,
      "epoch": 0.32445049601244896,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499314638768221,
      "loss": 5.6018,
      "mean_token_accuracy": 0.15770421475172042,
      "num_tokens": 7231853.0,
      "step": 4170
    },
    {
      "entropy": 5.761664533615113,
      "epoch": 0.3248395253841665,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.5293,
      "mean_token_accuracy": 0.16403157413005828,
      "num_tokens": 7241081.0,
      "step": 4175
    },
    {
      "entropy": 5.71790828704834,
      "epoch": 0.3252285547558841,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1609074741601944,
      "num_tokens": 7249742.0,
      "step": 4180
    },
    {
      "entropy": 5.8293393611907955,
      "epoch": 0.3256175841276016,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.6852,
      "mean_token_accuracy": 0.15708462446928023,
      "num_tokens": 7258285.0,
      "step": 4185
    },
    {
      "entropy": 5.750807237625122,
      "epoch": 0.3260066134993192,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.5018,
      "mean_token_accuracy": 0.16263256967067719,
      "num_tokens": 7266843.0,
      "step": 4190
    },
    {
      "entropy": 5.682654571533203,
      "epoch": 0.3263956428710368,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.5913,
      "mean_token_accuracy": 0.16370514035224915,
      "num_tokens": 7276041.0,
      "step": 4195
    },
    {
      "entropy": 5.738144969940185,
      "epoch": 0.3267846722427543,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.6502,
      "mean_token_accuracy": 0.15509369671344758,
      "num_tokens": 7285144.0,
      "step": 4200
    },
    {
      "entropy": 5.65037989616394,
      "epoch": 0.3271737016144719,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499299424136433,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16100571006536485,
      "num_tokens": 7293976.0,
      "step": 4205
    },
    {
      "entropy": 5.765503311157227,
      "epoch": 0.32756273098618943,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.5695,
      "mean_token_accuracy": 0.15914326161146164,
      "num_tokens": 7302365.0,
      "step": 4210
    },
    {
      "entropy": 5.791704416275024,
      "epoch": 0.327951760357907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.6453,
      "mean_token_accuracy": 0.15589653253555297,
      "num_tokens": 7310909.0,
      "step": 4215
    },
    {
      "entropy": 5.762666463851929,
      "epoch": 0.3283407897296246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.6041,
      "mean_token_accuracy": 0.15364096760749818,
      "num_tokens": 7319391.0,
      "step": 4220
    },
    {
      "entropy": 5.86359977722168,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499290655185891,
      "loss": 5.6782,
      "mean_token_accuracy": 0.16046358346939088,
      "num_tokens": 7328076.0,
      "step": 4225
    },
    {
      "entropy": 5.7980348587036135,
      "epoch": 0.3291188484730597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499288454440038,
      "loss": 5.6062,
      "mean_token_accuracy": 0.15722377076745034,
      "num_tokens": 7336833.0,
      "step": 4230
    },
    {
      "entropy": 5.711167240142823,
      "epoch": 0.3295078778447773,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.5223,
      "mean_token_accuracy": 0.1646954208612442,
      "num_tokens": 7344862.0,
      "step": 4235
    },
    {
      "entropy": 5.775886392593383,
      "epoch": 0.32989690721649484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.5541,
      "mean_token_accuracy": 0.16591128408908845,
      "num_tokens": 7354503.0,
      "step": 4240
    },
    {
      "entropy": 5.760515594482422,
      "epoch": 0.3302859365882124,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.5568,
      "mean_token_accuracy": 0.16480007916688919,
      "num_tokens": 7363000.0,
      "step": 4245
    },
    {
      "entropy": 5.701372575759888,
      "epoch": 0.33067496595992996,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.6312,
      "mean_token_accuracy": 0.15584687665104865,
      "num_tokens": 7372619.0,
      "step": 4250
    },
    {
      "entropy": 5.803398752212525,
      "epoch": 0.33106399533164754,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.5015,
      "mean_token_accuracy": 0.17060409784317015,
      "num_tokens": 7380531.0,
      "step": 4255
    },
    {
      "entropy": 5.745937347412109,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.6194,
      "mean_token_accuracy": 0.16034240871667862,
      "num_tokens": 7389153.0,
      "step": 4260
    },
    {
      "entropy": 5.725397109985352,
      "epoch": 0.33184205407508266,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.5189,
      "mean_token_accuracy": 0.16305640786886216,
      "num_tokens": 7397880.0,
      "step": 4265
    },
    {
      "entropy": 5.78955659866333,
      "epoch": 0.33223108344680025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.5854,
      "mean_token_accuracy": 0.15720919668674468,
      "num_tokens": 7406180.0,
      "step": 4270
    },
    {
      "entropy": 5.7014524936676025,
      "epoch": 0.3326201128185178,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.593,
      "mean_token_accuracy": 0.16094412356615068,
      "num_tokens": 7415069.0,
      "step": 4275
    },
    {
      "entropy": 5.744061803817749,
      "epoch": 0.33300914219023536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.6493,
      "mean_token_accuracy": 0.15555099695920943,
      "num_tokens": 7423921.0,
      "step": 4280
    },
    {
      "entropy": 5.696127796173096,
      "epoch": 0.33339817156195295,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.3854,
      "mean_token_accuracy": 0.17118807137012482,
      "num_tokens": 7432655.0,
      "step": 4285
    },
    {
      "entropy": 5.776129102706909,
      "epoch": 0.3337872009336705,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.578,
      "mean_token_accuracy": 0.16193160861730577,
      "num_tokens": 7441858.0,
      "step": 4290
    },
    {
      "entropy": 5.6602355480194095,
      "epoch": 0.33417623030538807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499259535063126,
      "loss": 5.5398,
      "mean_token_accuracy": 0.16309095025062562,
      "num_tokens": 7450030.0,
      "step": 4295
    },
    {
      "entropy": 5.785944986343384,
      "epoch": 0.3345652596771056,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499257286675251,
      "loss": 5.6258,
      "mean_token_accuracy": 0.15881484746932983,
      "num_tokens": 7458256.0,
      "step": 4300
    },
    {
      "entropy": 5.786884784698486,
      "epoch": 0.3349542890488232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.5915,
      "mean_token_accuracy": 0.15856461822986603,
      "num_tokens": 7466710.0,
      "step": 4305
    },
    {
      "entropy": 5.710636949539184,
      "epoch": 0.33534331842054077,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.5344,
      "mean_token_accuracy": 0.16328217089176178,
      "num_tokens": 7475445.0,
      "step": 4310
    },
    {
      "entropy": 5.79075436592102,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.6155,
      "mean_token_accuracy": 0.15852408558130265,
      "num_tokens": 7484326.0,
      "step": 4315
    },
    {
      "entropy": 5.715756559371949,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.5138,
      "mean_token_accuracy": 0.15974442958831786,
      "num_tokens": 7493028.0,
      "step": 4320
    },
    {
      "entropy": 5.810898685455323,
      "epoch": 0.3365104065356934,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.691,
      "mean_token_accuracy": 0.1529371842741966,
      "num_tokens": 7502284.0,
      "step": 4325
    },
    {
      "entropy": 5.789745283126831,
      "epoch": 0.336899435907411,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.572,
      "mean_token_accuracy": 0.15930989682674407,
      "num_tokens": 7510728.0,
      "step": 4330
    },
    {
      "entropy": 5.705639839172363,
      "epoch": 0.3372884652791286,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.5308,
      "mean_token_accuracy": 0.1664871394634247,
      "num_tokens": 7519578.0,
      "step": 4335
    },
    {
      "entropy": 5.7081053256988525,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.6538,
      "mean_token_accuracy": 0.15889098793268203,
      "num_tokens": 7528853.0,
      "step": 4340
    },
    {
      "entropy": 5.90105094909668,
      "epoch": 0.3380665240225637,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499236898064854,
      "loss": 5.6981,
      "mean_token_accuracy": 0.156148099899292,
      "num_tokens": 7537000.0,
      "step": 4345
    },
    {
      "entropy": 5.771493673324585,
      "epoch": 0.3384555533942813,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.6498,
      "mean_token_accuracy": 0.15839225202798843,
      "num_tokens": 7545439.0,
      "step": 4350
    },
    {
      "entropy": 5.763132333755493,
      "epoch": 0.3388445827659988,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.5485,
      "mean_token_accuracy": 0.1569911375641823,
      "num_tokens": 7553808.0,
      "step": 4355
    },
    {
      "entropy": 5.793811559677124,
      "epoch": 0.3392336121377164,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499230040616263,
      "loss": 5.5838,
      "mean_token_accuracy": 0.16720182895660402,
      "num_tokens": 7562124.0,
      "step": 4360
    },
    {
      "entropy": 5.779132938385009,
      "epoch": 0.33962264150943394,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.626,
      "mean_token_accuracy": 0.1591033309698105,
      "num_tokens": 7570692.0,
      "step": 4365
    },
    {
      "entropy": 5.69827675819397,
      "epoch": 0.34001167088115153,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1599597990512848,
      "num_tokens": 7579380.0,
      "step": 4370
    },
    {
      "entropy": 5.712707996368408,
      "epoch": 0.3404007002528691,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.5748,
      "mean_token_accuracy": 0.15763231217861176,
      "num_tokens": 7587581.0,
      "step": 4375
    },
    {
      "entropy": 5.803431415557862,
      "epoch": 0.34078972962458665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.6207,
      "mean_token_accuracy": 0.15856635421514512,
      "num_tokens": 7596373.0,
      "step": 4380
    },
    {
      "entropy": 5.769613647460938,
      "epoch": 0.34117875899630423,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.63,
      "mean_token_accuracy": 0.15952463150024415,
      "num_tokens": 7605698.0,
      "step": 4385
    },
    {
      "entropy": 5.831510257720947,
      "epoch": 0.34156778836802176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.6701,
      "mean_token_accuracy": 0.15391338169574736,
      "num_tokens": 7614299.0,
      "step": 4390
    },
    {
      "entropy": 5.711894607543945,
      "epoch": 0.34195681773973935,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1665854424238205,
      "num_tokens": 7622244.0,
      "step": 4395
    },
    {
      "entropy": 5.799953508377075,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.6835,
      "mean_token_accuracy": 0.14930743724107742,
      "num_tokens": 7631421.0,
      "step": 4400
    },
    {
      "entropy": 5.780383348464966,
      "epoch": 0.34273487648317447,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.617,
      "mean_token_accuracy": 0.1649213418364525,
      "num_tokens": 7640483.0,
      "step": 4405
    },
    {
      "entropy": 5.734336900711059,
      "epoch": 0.34312390585489205,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.5434,
      "mean_token_accuracy": 0.16003944873809814,
      "num_tokens": 7648928.0,
      "step": 4410
    },
    {
      "entropy": 5.748956537246704,
      "epoch": 0.3435129352266096,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.556,
      "mean_token_accuracy": 0.16345997750759125,
      "num_tokens": 7658158.0,
      "step": 4415
    },
    {
      "entropy": 5.79419469833374,
      "epoch": 0.34390196459832717,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.6394,
      "mean_token_accuracy": 0.15967277586460113,
      "num_tokens": 7666657.0,
      "step": 4420
    },
    {
      "entropy": 5.756088972091675,
      "epoch": 0.34429099397004476,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.5864,
      "mean_token_accuracy": 0.1604827433824539,
      "num_tokens": 7675182.0,
      "step": 4425
    },
    {
      "entropy": 5.713095378875733,
      "epoch": 0.3446800233417623,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.537,
      "mean_token_accuracy": 0.16038379073143005,
      "num_tokens": 7684132.0,
      "step": 4430
    },
    {
      "entropy": 5.707273960113525,
      "epoch": 0.3450690527134799,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.4828,
      "mean_token_accuracy": 0.16817560493946077,
      "num_tokens": 7692574.0,
      "step": 4435
    },
    {
      "entropy": 5.682709455490112,
      "epoch": 0.34545808208519746,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.5448,
      "mean_token_accuracy": 0.1668045476078987,
      "num_tokens": 7700969.0,
      "step": 4440
    },
    {
      "entropy": 5.786290264129638,
      "epoch": 0.345847111456915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15646209716796874,
      "num_tokens": 7710688.0,
      "step": 4445
    },
    {
      "entropy": 5.8290328025817875,
      "epoch": 0.3462361408286326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1635972410440445,
      "num_tokens": 7720071.0,
      "step": 4450
    },
    {
      "entropy": 5.673279428482056,
      "epoch": 0.3466251702003501,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.4371,
      "mean_token_accuracy": 0.16662000864744186,
      "num_tokens": 7728418.0,
      "step": 4455
    },
    {
      "entropy": 5.733144617080688,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.67,
      "mean_token_accuracy": 0.1602802112698555,
      "num_tokens": 7736849.0,
      "step": 4460
    },
    {
      "entropy": 5.787078762054444,
      "epoch": 0.3474032289437853,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.5981,
      "mean_token_accuracy": 0.1622970446944237,
      "num_tokens": 7745730.0,
      "step": 4465
    },
    {
      "entropy": 5.771909475326538,
      "epoch": 0.3477922583155028,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.4965,
      "mean_token_accuracy": 0.16542282104492187,
      "num_tokens": 7755040.0,
      "step": 4470
    },
    {
      "entropy": 5.734687471389771,
      "epoch": 0.3481812876872204,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.5207,
      "mean_token_accuracy": 0.16281888782978057,
      "num_tokens": 7763765.0,
      "step": 4475
    },
    {
      "entropy": 5.622528505325318,
      "epoch": 0.34857031705893793,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16116178408265114,
      "num_tokens": 7772261.0,
      "step": 4480
    },
    {
      "entropy": 5.827115106582641,
      "epoch": 0.3489593464306555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.6628,
      "mean_token_accuracy": 0.15989381670951844,
      "num_tokens": 7781651.0,
      "step": 4485
    },
    {
      "entropy": 5.714611434936524,
      "epoch": 0.3493483758023731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.518,
      "mean_token_accuracy": 0.15781033784151077,
      "num_tokens": 7789926.0,
      "step": 4490
    },
    {
      "entropy": 5.715775346755981,
      "epoch": 0.34973740517409063,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.5567,
      "mean_token_accuracy": 0.15968117415904998,
      "num_tokens": 7798435.0,
      "step": 4495
    },
    {
      "entropy": 5.662638235092163,
      "epoch": 0.3501264345458082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.4951,
      "mean_token_accuracy": 0.15639443546533585,
      "num_tokens": 7806674.0,
      "step": 4500
    },
    {
      "entropy": 5.754309892654419,
      "epoch": 0.35051546391752575,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.5444,
      "mean_token_accuracy": 0.16342764496803283,
      "num_tokens": 7814774.0,
      "step": 4505
    },
    {
      "entropy": 5.704548788070679,
      "epoch": 0.35090449328924334,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.5986,
      "mean_token_accuracy": 0.16580508202314376,
      "num_tokens": 7822879.0,
      "step": 4510
    },
    {
      "entropy": 5.696450519561767,
      "epoch": 0.3512935226609609,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.5416,
      "mean_token_accuracy": 0.15570010393857955,
      "num_tokens": 7831418.0,
      "step": 4515
    },
    {
      "entropy": 5.7250312805175785,
      "epoch": 0.35168255203267845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.5079,
      "mean_token_accuracy": 0.15775506496429442,
      "num_tokens": 7839513.0,
      "step": 4520
    },
    {
      "entropy": 5.691388750076294,
      "epoch": 0.35207158140439604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.5324,
      "mean_token_accuracy": 0.16139761209487916,
      "num_tokens": 7847653.0,
      "step": 4525
    },
    {
      "entropy": 5.782090950012207,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.6185,
      "mean_token_accuracy": 0.1575747847557068,
      "num_tokens": 7857093.0,
      "step": 4530
    },
    {
      "entropy": 5.737625694274902,
      "epoch": 0.35284964014783116,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.582,
      "mean_token_accuracy": 0.1566581904888153,
      "num_tokens": 7866129.0,
      "step": 4535
    },
    {
      "entropy": 5.70094780921936,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.6615,
      "mean_token_accuracy": 0.15456809997558593,
      "num_tokens": 7874613.0,
      "step": 4540
    },
    {
      "entropy": 5.6414001941680905,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.4164,
      "mean_token_accuracy": 0.16817961633205414,
      "num_tokens": 7882292.0,
      "step": 4545
    },
    {
      "entropy": 5.737434244155883,
      "epoch": 0.35401672826298386,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.4426,
      "mean_token_accuracy": 0.17088190019130706,
      "num_tokens": 7890810.0,
      "step": 4550
    },
    {
      "entropy": 5.730095767974854,
      "epoch": 0.35440575763470145,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15629213452339172,
      "num_tokens": 7899964.0,
      "step": 4555
    },
    {
      "entropy": 5.672501754760742,
      "epoch": 0.354794787006419,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.4754,
      "mean_token_accuracy": 0.16402625739574433,
      "num_tokens": 7908600.0,
      "step": 4560
    },
    {
      "entropy": 5.677447080612183,
      "epoch": 0.35518381637813656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.5912,
      "mean_token_accuracy": 0.15798643827438355,
      "num_tokens": 7916893.0,
      "step": 4565
    },
    {
      "entropy": 5.731996774673462,
      "epoch": 0.3555728457498541,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1716650128364563,
      "num_tokens": 7925230.0,
      "step": 4570
    },
    {
      "entropy": 5.719581079483032,
      "epoch": 0.3559618751215717,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.5052,
      "mean_token_accuracy": 0.16825538873672485,
      "num_tokens": 7933487.0,
      "step": 4575
    },
    {
      "entropy": 5.671692514419556,
      "epoch": 0.35635090449328927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.5453,
      "mean_token_accuracy": 0.16380455940961838,
      "num_tokens": 7941957.0,
      "step": 4580
    },
    {
      "entropy": 5.7017723560333256,
      "epoch": 0.3567399338650068,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.4838,
      "mean_token_accuracy": 0.16157088428735733,
      "num_tokens": 7950912.0,
      "step": 4585
    },
    {
      "entropy": 5.68826847076416,
      "epoch": 0.3571289632367244,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.5453,
      "mean_token_accuracy": 0.16106254905462264,
      "num_tokens": 7959724.0,
      "step": 4590
    },
    {
      "entropy": 5.739056968688965,
      "epoch": 0.3575179926084419,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.5182,
      "mean_token_accuracy": 0.1671526774764061,
      "num_tokens": 7968683.0,
      "step": 4595
    },
    {
      "entropy": 5.694250869750976,
      "epoch": 0.3579070219801595,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499116157631659,
      "loss": 5.4915,
      "mean_token_accuracy": 0.1625325247645378,
      "num_tokens": 7977004.0,
      "step": 4600
    },
    {
      "entropy": 5.774178075790405,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1508305624127388,
      "num_tokens": 7985537.0,
      "step": 4605
    },
    {
      "entropy": 5.810094594955444,
      "epoch": 0.3586850807235946,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.5914,
      "mean_token_accuracy": 0.16004215627908708,
      "num_tokens": 7994298.0,
      "step": 4610
    },
    {
      "entropy": 5.675488615036011,
      "epoch": 0.3590741100953122,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.5485,
      "mean_token_accuracy": 0.16430403739213945,
      "num_tokens": 8003145.0,
      "step": 4615
    },
    {
      "entropy": 5.653676176071167,
      "epoch": 0.35946313946702974,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.5177,
      "mean_token_accuracy": 0.16057982444763183,
      "num_tokens": 8011839.0,
      "step": 4620
    },
    {
      "entropy": 5.8171083450317385,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.5952,
      "mean_token_accuracy": 0.15525781512260436,
      "num_tokens": 8020248.0,
      "step": 4625
    },
    {
      "entropy": 5.770076322555542,
      "epoch": 0.3602411982104649,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.6226,
      "mean_token_accuracy": 0.1595271795988083,
      "num_tokens": 8029360.0,
      "step": 4630
    },
    {
      "entropy": 5.830290365219116,
      "epoch": 0.36063022758218244,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.6756,
      "mean_token_accuracy": 0.15414261519908906,
      "num_tokens": 8037407.0,
      "step": 4635
    },
    {
      "entropy": 5.773358583450317,
      "epoch": 0.3610192569539,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499096415289178,
      "loss": 5.4892,
      "mean_token_accuracy": 0.1608835220336914,
      "num_tokens": 8045935.0,
      "step": 4640
    },
    {
      "entropy": 5.71015191078186,
      "epoch": 0.3614082863256176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.6419,
      "mean_token_accuracy": 0.14878360480070113,
      "num_tokens": 8054640.0,
      "step": 4645
    },
    {
      "entropy": 5.802717781066894,
      "epoch": 0.36179731569733514,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.6215,
      "mean_token_accuracy": 0.15971666425466538,
      "num_tokens": 8063845.0,
      "step": 4650
    },
    {
      "entropy": 5.790019845962524,
      "epoch": 0.36218634506905273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.5653,
      "mean_token_accuracy": 0.15723626762628556,
      "num_tokens": 8072389.0,
      "step": 4655
    },
    {
      "entropy": 5.651761722564697,
      "epoch": 0.36257537444077026,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1628458485007286,
      "num_tokens": 8080683.0,
      "step": 4660
    },
    {
      "entropy": 5.626598167419433,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.469,
      "mean_token_accuracy": 0.1658550351858139,
      "num_tokens": 8088945.0,
      "step": 4665
    },
    {
      "entropy": 5.662598133087158,
      "epoch": 0.36335343318420543,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.528,
      "mean_token_accuracy": 0.16415927559137344,
      "num_tokens": 8097447.0,
      "step": 4670
    },
    {
      "entropy": 5.654489183425904,
      "epoch": 0.36374246255592296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.3741,
      "mean_token_accuracy": 0.18078008890151978,
      "num_tokens": 8105570.0,
      "step": 4675
    },
    {
      "entropy": 5.64568943977356,
      "epoch": 0.36413149192764055,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.5065,
      "mean_token_accuracy": 0.16293679773807526,
      "num_tokens": 8114842.0,
      "step": 4680
    },
    {
      "entropy": 5.6961933135986325,
      "epoch": 0.3645205212993581,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.4941,
      "mean_token_accuracy": 0.16237769424915313,
      "num_tokens": 8122960.0,
      "step": 4685
    },
    {
      "entropy": 5.800394201278687,
      "epoch": 0.36490955067107567,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.6298,
      "mean_token_accuracy": 0.16036661416292192,
      "num_tokens": 8131850.0,
      "step": 4690
    },
    {
      "entropy": 5.730320596694947,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499068914240224,
      "loss": 5.5025,
      "mean_token_accuracy": 0.16529248505830765,
      "num_tokens": 8139818.0,
      "step": 4695
    },
    {
      "entropy": 5.6371876239776615,
      "epoch": 0.3656876094145108,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.5769,
      "mean_token_accuracy": 0.16349330991506578,
      "num_tokens": 8147995.0,
      "step": 4700
    },
    {
      "entropy": 5.758992433547974,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.5741,
      "mean_token_accuracy": 0.16156561821699142,
      "num_tokens": 8156879.0,
      "step": 4705
    },
    {
      "entropy": 5.740928316116333,
      "epoch": 0.3664656681579459,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.5433,
      "mean_token_accuracy": 0.150849449634552,
      "num_tokens": 8164860.0,
      "step": 4710
    },
    {
      "entropy": 5.624187231063843,
      "epoch": 0.3668546975296635,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17311475276947022,
      "num_tokens": 8173319.0,
      "step": 4715
    },
    {
      "entropy": 5.68842396736145,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.5738,
      "mean_token_accuracy": 0.16338748335838318,
      "num_tokens": 8182385.0,
      "step": 4720
    },
    {
      "entropy": 5.749617624282837,
      "epoch": 0.3676327562730986,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.494,
      "mean_token_accuracy": 0.16597676873207093,
      "num_tokens": 8190394.0,
      "step": 4725
    },
    {
      "entropy": 5.677096652984619,
      "epoch": 0.3680217856448162,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.5141,
      "mean_token_accuracy": 0.156700736284256,
      "num_tokens": 8198532.0,
      "step": 4730
    },
    {
      "entropy": 5.582213068008423,
      "epoch": 0.3684108150165337,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.4755,
      "mean_token_accuracy": 0.16821785122156144,
      "num_tokens": 8207436.0,
      "step": 4735
    },
    {
      "entropy": 5.749109268188477,
      "epoch": 0.3687998443882513,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.5093,
      "mean_token_accuracy": 0.16389241069555283,
      "num_tokens": 8216316.0,
      "step": 4740
    },
    {
      "entropy": 5.682870817184448,
      "epoch": 0.3691888737599689,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.4634,
      "mean_token_accuracy": 0.16937032341957092,
      "num_tokens": 8224918.0,
      "step": 4745
    },
    {
      "entropy": 5.652633953094482,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.4855,
      "mean_token_accuracy": 0.16341211497783661,
      "num_tokens": 8233506.0,
      "step": 4750
    },
    {
      "entropy": 5.686648225784301,
      "epoch": 0.369966932503404,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.5371,
      "mean_token_accuracy": 0.16136658936738968,
      "num_tokens": 8243122.0,
      "step": 4755
    },
    {
      "entropy": 5.695843458175659,
      "epoch": 0.3703559618751216,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.4954,
      "mean_token_accuracy": 0.16794812679290771,
      "num_tokens": 8251639.0,
      "step": 4760
    },
    {
      "entropy": 5.694685459136963,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.5441,
      "mean_token_accuracy": 0.1565728172659874,
      "num_tokens": 8260487.0,
      "step": 4765
    },
    {
      "entropy": 5.731024360656738,
      "epoch": 0.3711340206185567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.5353,
      "mean_token_accuracy": 0.16720337867736818,
      "num_tokens": 8270111.0,
      "step": 4770
    },
    {
      "entropy": 5.732585525512695,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.6699,
      "mean_token_accuracy": 0.15745235085487366,
      "num_tokens": 8278171.0,
      "step": 4775
    },
    {
      "entropy": 5.728489589691162,
      "epoch": 0.37191207936199183,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.5444,
      "mean_token_accuracy": 0.16361308097839355,
      "num_tokens": 8287278.0,
      "step": 4780
    },
    {
      "entropy": 5.801375198364258,
      "epoch": 0.3723011087337094,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.6581,
      "mean_token_accuracy": 0.15506795942783355,
      "num_tokens": 8296874.0,
      "step": 4785
    },
    {
      "entropy": 5.677485656738281,
      "epoch": 0.37269013810542695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.5292,
      "mean_token_accuracy": 0.16247798800468444,
      "num_tokens": 8305375.0,
      "step": 4790
    },
    {
      "entropy": 5.805545330047607,
      "epoch": 0.37307916747714454,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.5798,
      "mean_token_accuracy": 0.1593550831079483,
      "num_tokens": 8313960.0,
      "step": 4795
    },
    {
      "entropy": 5.732056093215943,
      "epoch": 0.37346819684886207,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.5091,
      "mean_token_accuracy": 0.1695062816143036,
      "num_tokens": 8321838.0,
      "step": 4800
    },
    {
      "entropy": 5.647253465652466,
      "epoch": 0.37385722622057965,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.549,
      "mean_token_accuracy": 0.15523339807987213,
      "num_tokens": 8331325.0,
      "step": 4805
    },
    {
      "entropy": 5.7562426090240475,
      "epoch": 0.37424625559229724,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.5963,
      "mean_token_accuracy": 0.1604474112391472,
      "num_tokens": 8340105.0,
      "step": 4810
    },
    {
      "entropy": 5.675045204162598,
      "epoch": 0.37463528496401477,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.4176,
      "mean_token_accuracy": 0.17652603089809418,
      "num_tokens": 8348256.0,
      "step": 4815
    },
    {
      "entropy": 5.65481014251709,
      "epoch": 0.37502431433573236,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.6006,
      "mean_token_accuracy": 0.15629589855670928,
      "num_tokens": 8357456.0,
      "step": 4820
    },
    {
      "entropy": 5.74207820892334,
      "epoch": 0.3754133437074499,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.4483,
      "mean_token_accuracy": 0.16962354630231857,
      "num_tokens": 8366939.0,
      "step": 4825
    },
    {
      "entropy": 5.838831901550293,
      "epoch": 0.3758023730791675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498999667628178,
      "loss": 5.6137,
      "mean_token_accuracy": 0.15240543484687805,
      "num_tokens": 8376062.0,
      "step": 4830
    },
    {
      "entropy": 5.771680784225464,
      "epoch": 0.37619140245088506,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.5276,
      "mean_token_accuracy": 0.16518781930208207,
      "num_tokens": 8383803.0,
      "step": 4835
    },
    {
      "entropy": 5.6745326042175295,
      "epoch": 0.3765804318226026,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.6889,
      "mean_token_accuracy": 0.15957241952419282,
      "num_tokens": 8392844.0,
      "step": 4840
    },
    {
      "entropy": 5.768671894073487,
      "epoch": 0.3769694611943202,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.5881,
      "mean_token_accuracy": 0.1660487473011017,
      "num_tokens": 8401937.0,
      "step": 4845
    },
    {
      "entropy": 5.746508407592773,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.5918,
      "mean_token_accuracy": 0.15559911131858825,
      "num_tokens": 8410383.0,
      "step": 4850
    },
    {
      "entropy": 5.664513111114502,
      "epoch": 0.3777475199377553,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.547,
      "mean_token_accuracy": 0.1587306007742882,
      "num_tokens": 8419053.0,
      "step": 4855
    },
    {
      "entropy": 5.702281188964844,
      "epoch": 0.3781365493094729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.5249,
      "mean_token_accuracy": 0.16742938309907912,
      "num_tokens": 8427311.0,
      "step": 4860
    },
    {
      "entropy": 5.715058994293213,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498981310309936,
      "loss": 5.5741,
      "mean_token_accuracy": 0.15908973962068557,
      "num_tokens": 8436050.0,
      "step": 4865
    },
    {
      "entropy": 5.820055341720581,
      "epoch": 0.378914608052908,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1646547108888626,
      "num_tokens": 8444579.0,
      "step": 4870
    },
    {
      "entropy": 5.689808559417725,
      "epoch": 0.3793036374246256,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.5276,
      "mean_token_accuracy": 0.1638904482126236,
      "num_tokens": 8454013.0,
      "step": 4875
    },
    {
      "entropy": 5.724506807327271,
      "epoch": 0.3796926667963431,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.6429,
      "mean_token_accuracy": 0.15735770165920257,
      "num_tokens": 8462564.0,
      "step": 4880
    },
    {
      "entropy": 5.7334497451782225,
      "epoch": 0.3800816961680607,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.4841,
      "mean_token_accuracy": 0.16679838299751282,
      "num_tokens": 8471035.0,
      "step": 4885
    },
    {
      "entropy": 5.680740404129028,
      "epoch": 0.38047072553977823,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.4693,
      "mean_token_accuracy": 0.16390866339206694,
      "num_tokens": 8479475.0,
      "step": 4890
    },
    {
      "entropy": 5.623473501205444,
      "epoch": 0.3808597549114958,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1630413204431534,
      "num_tokens": 8488163.0,
      "step": 4895
    },
    {
      "entropy": 5.821042728424072,
      "epoch": 0.3812487842832134,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.5995,
      "mean_token_accuracy": 0.15810228884220123,
      "num_tokens": 8497735.0,
      "step": 4900
    },
    {
      "entropy": 5.743423938751221,
      "epoch": 0.38163781365493094,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.5459,
      "mean_token_accuracy": 0.16006243228912354,
      "num_tokens": 8506986.0,
      "step": 4905
    },
    {
      "entropy": 5.6776947498321535,
      "epoch": 0.3820268430266485,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.5709,
      "mean_token_accuracy": 0.16254400610923767,
      "num_tokens": 8515168.0,
      "step": 4910
    },
    {
      "entropy": 5.690753602981568,
      "epoch": 0.38241587239836605,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.5509,
      "mean_token_accuracy": 0.16618261039257048,
      "num_tokens": 8523980.0,
      "step": 4915
    },
    {
      "entropy": 5.658663177490235,
      "epoch": 0.38280490177008364,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498952126643443,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16129754930734636,
      "num_tokens": 8532938.0,
      "step": 4920
    },
    {
      "entropy": 5.67104868888855,
      "epoch": 0.3831939311418012,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.4671,
      "mean_token_accuracy": 0.16780146807432175,
      "num_tokens": 8541951.0,
      "step": 4925
    },
    {
      "entropy": 5.74264087677002,
      "epoch": 0.38358296051351876,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.6436,
      "mean_token_accuracy": 0.15836040526628495,
      "num_tokens": 8550981.0,
      "step": 4930
    },
    {
      "entropy": 5.670666360855103,
      "epoch": 0.38397198988523634,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.5569,
      "mean_token_accuracy": 0.15649997144937516,
      "num_tokens": 8559711.0,
      "step": 4935
    },
    {
      "entropy": 5.681796836853027,
      "epoch": 0.3843610192569539,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498941412454744,
      "loss": 5.4853,
      "mean_token_accuracy": 0.15973234325647354,
      "num_tokens": 8567797.0,
      "step": 4940
    },
    {
      "entropy": 5.7254406929016115,
      "epoch": 0.38475004862867146,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.5791,
      "mean_token_accuracy": 0.15818444043397903,
      "num_tokens": 8576719.0,
      "step": 4945
    },
    {
      "entropy": 5.737820959091186,
      "epoch": 0.38513907800038905,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.5682,
      "mean_token_accuracy": 0.1667224258184433,
      "num_tokens": 8585871.0,
      "step": 4950
    },
    {
      "entropy": 5.708989763259888,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16250301152467728,
      "num_tokens": 8594018.0,
      "step": 4955
    },
    {
      "entropy": 5.680680894851685,
      "epoch": 0.38591713674382416,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.5976,
      "mean_token_accuracy": 0.15720874518156053,
      "num_tokens": 8602788.0,
      "step": 4960
    },
    {
      "entropy": 5.760657072067261,
      "epoch": 0.38630616611554175,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.5509,
      "mean_token_accuracy": 0.16338287740945817,
      "num_tokens": 8611837.0,
      "step": 4965
    },
    {
      "entropy": 5.704029893875122,
      "epoch": 0.3866951954872593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.5267,
      "mean_token_accuracy": 0.16973906457424165,
      "num_tokens": 8620114.0,
      "step": 4970
    },
    {
      "entropy": 5.8096014022827145,
      "epoch": 0.38708422485897687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.6419,
      "mean_token_accuracy": 0.15283987894654275,
      "num_tokens": 8628777.0,
      "step": 4975
    },
    {
      "entropy": 5.7505690574646,
      "epoch": 0.3874732542306944,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.5492,
      "mean_token_accuracy": 0.15922651141881944,
      "num_tokens": 8637088.0,
      "step": 4980
    },
    {
      "entropy": 5.732375717163086,
      "epoch": 0.387862283602412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.5598,
      "mean_token_accuracy": 0.16299061924219133,
      "num_tokens": 8645261.0,
      "step": 4985
    },
    {
      "entropy": 5.73756160736084,
      "epoch": 0.38825131297412957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498914389131909,
      "loss": 5.5689,
      "mean_token_accuracy": 0.16376165598630904,
      "num_tokens": 8653492.0,
      "step": 4990
    },
    {
      "entropy": 5.660993385314941,
      "epoch": 0.3886403423458471,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.5167,
      "mean_token_accuracy": 0.16651235818862914,
      "num_tokens": 8661909.0,
      "step": 4995
    },
    {
      "entropy": 5.676572227478028,
      "epoch": 0.3890293717175647,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.4822,
      "mean_token_accuracy": 0.16759965121746062,
      "num_tokens": 8670332.0,
      "step": 5000
    },
    {
      "entropy": 5.762911462783814,
      "epoch": 0.3894184010892822,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.5234,
      "mean_token_accuracy": 0.1680043324828148,
      "num_tokens": 8678537.0,
      "step": 5005
    },
    {
      "entropy": 5.692707681655884,
      "epoch": 0.3898074304609998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.4863,
      "mean_token_accuracy": 0.1656753972172737,
      "num_tokens": 8686839.0,
      "step": 5010
    },
    {
      "entropy": 5.6931657791137695,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.5426,
      "mean_token_accuracy": 0.16253531575202942,
      "num_tokens": 8694407.0,
      "step": 5015
    },
    {
      "entropy": 5.662943887710571,
      "epoch": 0.3905854892044349,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.4984,
      "mean_token_accuracy": 0.1695681780576706,
      "num_tokens": 8703484.0,
      "step": 5020
    },
    {
      "entropy": 5.807813596725464,
      "epoch": 0.3909745185761525,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498895270649594,
      "loss": 5.6039,
      "mean_token_accuracy": 0.1552526980638504,
      "num_tokens": 8712634.0,
      "step": 5025
    },
    {
      "entropy": 5.683317804336548,
      "epoch": 0.39136354794787004,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498892525848485,
      "loss": 5.487,
      "mean_token_accuracy": 0.1599026545882225,
      "num_tokens": 8721229.0,
      "step": 5030
    },
    {
      "entropy": 5.661917400360108,
      "epoch": 0.3917525773195876,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.541,
      "mean_token_accuracy": 0.16514813005924225,
      "num_tokens": 8730010.0,
      "step": 5035
    },
    {
      "entropy": 5.674532079696656,
      "epoch": 0.3921416066913052,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.5342,
      "mean_token_accuracy": 0.1578855812549591,
      "num_tokens": 8739046.0,
      "step": 5040
    },
    {
      "entropy": 5.742646265029907,
      "epoch": 0.39253063606302274,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.6198,
      "mean_token_accuracy": 0.16325782537460326,
      "num_tokens": 8747650.0,
      "step": 5045
    },
    {
      "entropy": 5.667177534103393,
      "epoch": 0.39291966543474033,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.5398,
      "mean_token_accuracy": 0.16407661586999894,
      "num_tokens": 8756319.0,
      "step": 5050
    },
    {
      "entropy": 5.6189886093139645,
      "epoch": 0.3933086948064579,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.554,
      "mean_token_accuracy": 0.16317882537841796,
      "num_tokens": 8765448.0,
      "step": 5055
    },
    {
      "entropy": 5.711760187149048,
      "epoch": 0.39369772417817545,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.5024,
      "mean_token_accuracy": 0.15837739109992982,
      "num_tokens": 8774146.0,
      "step": 5060
    },
    {
      "entropy": 5.642172765731812,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.3808,
      "mean_token_accuracy": 0.16725590229034423,
      "num_tokens": 8782311.0,
      "step": 5065
    },
    {
      "entropy": 5.620117568969727,
      "epoch": 0.39447578292161056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.4233,
      "mean_token_accuracy": 0.167247074842453,
      "num_tokens": 8790196.0,
      "step": 5070
    },
    {
      "entropy": 5.670018005371094,
      "epoch": 0.39486481229332815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.5098,
      "mean_token_accuracy": 0.1597319647669792,
      "num_tokens": 8798836.0,
      "step": 5075
    },
    {
      "entropy": 5.671954202651977,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498864890997593,
      "loss": 5.5441,
      "mean_token_accuracy": 0.16164022088050842,
      "num_tokens": 8806942.0,
      "step": 5080
    },
    {
      "entropy": 5.761148405075073,
      "epoch": 0.39564287103676327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.7312,
      "mean_token_accuracy": 0.15544241815805435,
      "num_tokens": 8816354.0,
      "step": 5085
    },
    {
      "entropy": 5.7759716510772705,
      "epoch": 0.39603190040848085,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498859323264548,
      "loss": 5.4963,
      "mean_token_accuracy": 0.1662037640810013,
      "num_tokens": 8824802.0,
      "step": 5090
    },
    {
      "entropy": 5.677498435974121,
      "epoch": 0.3964209297801984,
      "grad_norm": 1.0,
      "learning_rate": 0.000498856534302941,
      "loss": 5.4777,
      "mean_token_accuracy": 0.1677809789776802,
      "num_tokens": 8833348.0,
      "step": 5095
    },
    {
      "entropy": 5.632452917098999,
      "epoch": 0.39680995915191597,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16559952795505523,
      "num_tokens": 8841131.0,
      "step": 5100
    },
    {
      "entropy": 5.647958040237427,
      "epoch": 0.39719898852363356,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.4972,
      "mean_token_accuracy": 0.16812338382005693,
      "num_tokens": 8849768.0,
      "step": 5105
    },
    {
      "entropy": 5.73589940071106,
      "epoch": 0.3975880178953511,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.466,
      "mean_token_accuracy": 0.17053375840187074,
      "num_tokens": 8858170.0,
      "step": 5110
    },
    {
      "entropy": 5.652495384216309,
      "epoch": 0.3979770472670687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.4898,
      "mean_token_accuracy": 0.16427329331636428,
      "num_tokens": 8866720.0,
      "step": 5115
    },
    {
      "entropy": 5.744022846221924,
      "epoch": 0.3983660766387862,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.5027,
      "mean_token_accuracy": 0.1700969159603119,
      "num_tokens": 8875141.0,
      "step": 5120
    },
    {
      "entropy": 5.721825456619262,
      "epoch": 0.3987551060105038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.5215,
      "mean_token_accuracy": 0.17009811848402023,
      "num_tokens": 8884512.0,
      "step": 5125
    },
    {
      "entropy": 5.744118022918701,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.5924,
      "mean_token_accuracy": 0.1634611889719963,
      "num_tokens": 8893487.0,
      "step": 5130
    },
    {
      "entropy": 5.661117696762085,
      "epoch": 0.3995331647539389,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.5359,
      "mean_token_accuracy": 0.1611662432551384,
      "num_tokens": 8902909.0,
      "step": 5135
    },
    {
      "entropy": 5.72418098449707,
      "epoch": 0.3999221941256565,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.4443,
      "mean_token_accuracy": 0.1671905130147934,
      "num_tokens": 8911311.0,
      "step": 5140
    },
    {
      "entropy": 5.624884366989136,
      "epoch": 0.400311223497374,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.4728,
      "mean_token_accuracy": 0.1723196566104889,
      "num_tokens": 8920222.0,
      "step": 5145
    },
    {
      "entropy": 5.642570638656617,
      "epoch": 0.4007002528690916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.5248,
      "mean_token_accuracy": 0.16120615154504775,
      "num_tokens": 8928763.0,
      "step": 5150
    },
    {
      "entropy": 5.795141744613647,
      "epoch": 0.4010892822408092,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.5824,
      "mean_token_accuracy": 0.16073987632989883,
      "num_tokens": 8937832.0,
      "step": 5155
    },
    {
      "entropy": 5.649924993515015,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.4415,
      "mean_token_accuracy": 0.1690294101834297,
      "num_tokens": 8945662.0,
      "step": 5160
    },
    {
      "entropy": 5.631535243988037,
      "epoch": 0.4018673409842443,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.5209,
      "mean_token_accuracy": 0.16340990066528321,
      "num_tokens": 8954570.0,
      "step": 5165
    },
    {
      "entropy": 5.723238515853882,
      "epoch": 0.4022563703559619,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.5174,
      "mean_token_accuracy": 0.16258451640605925,
      "num_tokens": 8962827.0,
      "step": 5170
    },
    {
      "entropy": 5.70035228729248,
      "epoch": 0.40264539972767943,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.5363,
      "mean_token_accuracy": 0.16639768928289414,
      "num_tokens": 8972119.0,
      "step": 5175
    },
    {
      "entropy": 5.6737593650817875,
      "epoch": 0.403034429099397,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16658223271369935,
      "num_tokens": 8980498.0,
      "step": 5180
    },
    {
      "entropy": 5.702155590057373,
      "epoch": 0.40342345847111455,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.6045,
      "mean_token_accuracy": 0.16114473044872285,
      "num_tokens": 8989819.0,
      "step": 5185
    },
    {
      "entropy": 5.76625337600708,
      "epoch": 0.40381248784283214,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.6453,
      "mean_token_accuracy": 0.15599432587623596,
      "num_tokens": 8998884.0,
      "step": 5190
    },
    {
      "entropy": 5.686150741577149,
      "epoch": 0.4042015172145497,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.5014,
      "mean_token_accuracy": 0.1671816885471344,
      "num_tokens": 9007806.0,
      "step": 5195
    },
    {
      "entropy": 5.734596157073975,
      "epoch": 0.40459054658626725,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.6647,
      "mean_token_accuracy": 0.15147211849689485,
      "num_tokens": 9016541.0,
      "step": 5200
    },
    {
      "entropy": 5.6630021095275875,
      "epoch": 0.40497957595798484,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.429,
      "mean_token_accuracy": 0.16656740009784698,
      "num_tokens": 9024503.0,
      "step": 5205
    },
    {
      "entropy": 5.718490028381348,
      "epoch": 0.40536860532970237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.5085,
      "mean_token_accuracy": 0.16654182076454163,
      "num_tokens": 9033182.0,
      "step": 5210
    },
    {
      "entropy": 5.655713081359863,
      "epoch": 0.40575763470141996,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.5342,
      "mean_token_accuracy": 0.1643572121858597,
      "num_tokens": 9042281.0,
      "step": 5215
    },
    {
      "entropy": 5.745838260650634,
      "epoch": 0.40614666407313754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.447,
      "mean_token_accuracy": 0.1690574824810028,
      "num_tokens": 9050570.0,
      "step": 5220
    },
    {
      "entropy": 5.760504627227784,
      "epoch": 0.4065356934448551,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1569422021508217,
      "num_tokens": 9059782.0,
      "step": 5225
    },
    {
      "entropy": 5.624711227416992,
      "epoch": 0.40692472281657266,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498779948539433,
      "loss": 5.3755,
      "mean_token_accuracy": 0.1710534617304802,
      "num_tokens": 9067484.0,
      "step": 5230
    },
    {
      "entropy": 5.667460870742798,
      "epoch": 0.4073137521882902,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.5824,
      "mean_token_accuracy": 0.15947526395320893,
      "num_tokens": 9075740.0,
      "step": 5235
    },
    {
      "entropy": 5.768729782104492,
      "epoch": 0.4077027815600078,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.6286,
      "mean_token_accuracy": 0.15769142508506775,
      "num_tokens": 9084137.0,
      "step": 5240
    },
    {
      "entropy": 5.665251207351685,
      "epoch": 0.40809181093172536,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.4091,
      "mean_token_accuracy": 0.17634232938289643,
      "num_tokens": 9093245.0,
      "step": 5245
    },
    {
      "entropy": 5.578675079345703,
      "epoch": 0.4084808403034429,
      "grad_norm": 1.0,
      "learning_rate": 0.000498768391958494,
      "loss": 5.4075,
      "mean_token_accuracy": 0.16961877197027206,
      "num_tokens": 9102837.0,
      "step": 5250
    },
    {
      "entropy": 5.6759144306182865,
      "epoch": 0.4088698696751605,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.4679,
      "mean_token_accuracy": 0.16916224360466003,
      "num_tokens": 9111087.0,
      "step": 5255
    },
    {
      "entropy": 5.666718244552612,
      "epoch": 0.40925889904687807,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.4955,
      "mean_token_accuracy": 0.17213426232337953,
      "num_tokens": 9120506.0,
      "step": 5260
    },
    {
      "entropy": 5.754262399673462,
      "epoch": 0.4096479284185956,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.6924,
      "mean_token_accuracy": 0.15665965899825096,
      "num_tokens": 9129637.0,
      "step": 5265
    },
    {
      "entropy": 5.753048419952393,
      "epoch": 0.4100369577903132,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.4725,
      "mean_token_accuracy": 0.16619387418031692,
      "num_tokens": 9139338.0,
      "step": 5270
    },
    {
      "entropy": 5.6472631931304935,
      "epoch": 0.4104259871620307,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.5164,
      "mean_token_accuracy": 0.16488752961158754,
      "num_tokens": 9148034.0,
      "step": 5275
    },
    {
      "entropy": 5.662676572799683,
      "epoch": 0.4108150165337483,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.4633,
      "mean_token_accuracy": 0.16608307063579558,
      "num_tokens": 9156744.0,
      "step": 5280
    },
    {
      "entropy": 5.606441593170166,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498748037223375,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1717175617814064,
      "num_tokens": 9165174.0,
      "step": 5285
    },
    {
      "entropy": 5.640111970901489,
      "epoch": 0.4115930752771834,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.451,
      "mean_token_accuracy": 0.17056777477264404,
      "num_tokens": 9173945.0,
      "step": 5290
    },
    {
      "entropy": 5.71455397605896,
      "epoch": 0.411982104648901,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.4518,
      "mean_token_accuracy": 0.16707825809717178,
      "num_tokens": 9182728.0,
      "step": 5295
    },
    {
      "entropy": 5.591234254837036,
      "epoch": 0.41237113402061853,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1636035606265068,
      "num_tokens": 9191494.0,
      "step": 5300
    },
    {
      "entropy": 5.655320119857788,
      "epoch": 0.4127601633923361,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.5143,
      "mean_token_accuracy": 0.16841454058885574,
      "num_tokens": 9199190.0,
      "step": 5305
    },
    {
      "entropy": 5.651890897750855,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.4463,
      "mean_token_accuracy": 0.16756607592105865,
      "num_tokens": 9208066.0,
      "step": 5310
    },
    {
      "entropy": 5.643588495254517,
      "epoch": 0.41353822213577124,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1683342546224594,
      "num_tokens": 9217103.0,
      "step": 5315
    },
    {
      "entropy": 5.648463869094849,
      "epoch": 0.4139272515074888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.4995,
      "mean_token_accuracy": 0.16448337435722352,
      "num_tokens": 9225786.0,
      "step": 5320
    },
    {
      "entropy": 5.739774513244629,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.5035,
      "mean_token_accuracy": 0.171517077088356,
      "num_tokens": 9234182.0,
      "step": 5325
    },
    {
      "entropy": 5.693743753433227,
      "epoch": 0.41470531025092394,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1606791913509369,
      "num_tokens": 9243844.0,
      "step": 5330
    },
    {
      "entropy": 5.649951553344726,
      "epoch": 0.41509433962264153,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.4631,
      "mean_token_accuracy": 0.17001460939645768,
      "num_tokens": 9252901.0,
      "step": 5335
    },
    {
      "entropy": 5.588115978240967,
      "epoch": 0.41548336899435906,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.4564,
      "mean_token_accuracy": 0.16492853313684464,
      "num_tokens": 9260947.0,
      "step": 5340
    },
    {
      "entropy": 5.7257404804229735,
      "epoch": 0.41587239836607665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1641767904162407,
      "num_tokens": 9269249.0,
      "step": 5345
    },
    {
      "entropy": 5.650511598587036,
      "epoch": 0.4162614277377942,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.5138,
      "mean_token_accuracy": 0.16471074074506759,
      "num_tokens": 9277668.0,
      "step": 5350
    },
    {
      "entropy": 5.667751789093018,
      "epoch": 0.41665045710951176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.5103,
      "mean_token_accuracy": 0.15986161753535272,
      "num_tokens": 9286000.0,
      "step": 5355
    },
    {
      "entropy": 5.684542274475097,
      "epoch": 0.41703948648122935,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.5558,
      "mean_token_accuracy": 0.15900709629058837,
      "num_tokens": 9295090.0,
      "step": 5360
    },
    {
      "entropy": 5.656639814376831,
      "epoch": 0.4174285158529469,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.4706,
      "mean_token_accuracy": 0.16817761659622193,
      "num_tokens": 9302996.0,
      "step": 5365
    },
    {
      "entropy": 5.613350296020508,
      "epoch": 0.41781754522466447,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.3777,
      "mean_token_accuracy": 0.17125458717346193,
      "num_tokens": 9311636.0,
      "step": 5370
    },
    {
      "entropy": 5.645034265518189,
      "epoch": 0.41820657459638205,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.4855,
      "mean_token_accuracy": 0.16353526413440705,
      "num_tokens": 9319884.0,
      "step": 5375
    },
    {
      "entropy": 5.59897813796997,
      "epoch": 0.4185956039680996,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.4166,
      "mean_token_accuracy": 0.16931099370121955,
      "num_tokens": 9328273.0,
      "step": 5380
    },
    {
      "entropy": 5.65986647605896,
      "epoch": 0.41898463333981717,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1573350176215172,
      "num_tokens": 9337696.0,
      "step": 5385
    },
    {
      "entropy": 5.6967016696929935,
      "epoch": 0.4193736627115347,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16329908072948457,
      "num_tokens": 9346596.0,
      "step": 5390
    },
    {
      "entropy": 5.662880325317383,
      "epoch": 0.4197626920832523,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16471709311008453,
      "num_tokens": 9355480.0,
      "step": 5395
    },
    {
      "entropy": 5.650768995285034,
      "epoch": 0.4201517214549699,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.5032,
      "mean_token_accuracy": 0.166390161216259,
      "num_tokens": 9364422.0,
      "step": 5400
    },
    {
      "entropy": 5.634923124313355,
      "epoch": 0.4205407508266874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.3819,
      "mean_token_accuracy": 0.16537950336933135,
      "num_tokens": 9372311.0,
      "step": 5405
    },
    {
      "entropy": 5.655995321273804,
      "epoch": 0.420929780198405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.5105,
      "mean_token_accuracy": 0.16196997761726378,
      "num_tokens": 9380936.0,
      "step": 5410
    },
    {
      "entropy": 5.6495442390441895,
      "epoch": 0.4213188095701225,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.4565,
      "mean_token_accuracy": 0.16568768620491028,
      "num_tokens": 9390008.0,
      "step": 5415
    },
    {
      "entropy": 5.68368182182312,
      "epoch": 0.4217078389418401,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.539,
      "mean_token_accuracy": 0.16439356058835983,
      "num_tokens": 9398771.0,
      "step": 5420
    },
    {
      "entropy": 5.65579195022583,
      "epoch": 0.4220968683135577,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.5314,
      "mean_token_accuracy": 0.16600719690322877,
      "num_tokens": 9407105.0,
      "step": 5425
    },
    {
      "entropy": 5.6679205894470215,
      "epoch": 0.4224858976852752,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1649916112422943,
      "num_tokens": 9415737.0,
      "step": 5430
    },
    {
      "entropy": 5.6381096839904785,
      "epoch": 0.4228749270569928,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16958317011594773,
      "num_tokens": 9424661.0,
      "step": 5435
    },
    {
      "entropy": 5.731217908859253,
      "epoch": 0.42326395642871034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.535,
      "mean_token_accuracy": 0.1611144796013832,
      "num_tokens": 9433971.0,
      "step": 5440
    },
    {
      "entropy": 5.628467655181884,
      "epoch": 0.42365298580042793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.4786,
      "mean_token_accuracy": 0.16527963280677796,
      "num_tokens": 9442905.0,
      "step": 5445
    },
    {
      "entropy": 5.631464242935181,
      "epoch": 0.4240420151721455,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.5061,
      "mean_token_accuracy": 0.1702937290072441,
      "num_tokens": 9451258.0,
      "step": 5450
    },
    {
      "entropy": 5.669587421417236,
      "epoch": 0.42443104454386305,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16737881749868394,
      "num_tokens": 9460218.0,
      "step": 5455
    },
    {
      "entropy": 5.6664165496826175,
      "epoch": 0.42482007391558063,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.5074,
      "mean_token_accuracy": 0.1663854479789734,
      "num_tokens": 9468756.0,
      "step": 5460
    },
    {
      "entropy": 5.695584869384765,
      "epoch": 0.4252091032872982,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.528,
      "mean_token_accuracy": 0.16710674911737441,
      "num_tokens": 9477622.0,
      "step": 5465
    },
    {
      "entropy": 5.6477516174316404,
      "epoch": 0.42559813265901575,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.4023,
      "mean_token_accuracy": 0.17134957611560822,
      "num_tokens": 9486332.0,
      "step": 5470
    },
    {
      "entropy": 5.6369672298431395,
      "epoch": 0.42598716203073334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1770221471786499,
      "num_tokens": 9494703.0,
      "step": 5475
    },
    {
      "entropy": 5.6113409996032715,
      "epoch": 0.42637619140245087,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.3682,
      "mean_token_accuracy": 0.17947474718093873,
      "num_tokens": 9503008.0,
      "step": 5480
    },
    {
      "entropy": 5.647426748275757,
      "epoch": 0.42676522077416845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.475,
      "mean_token_accuracy": 0.16509030908346176,
      "num_tokens": 9511752.0,
      "step": 5485
    },
    {
      "entropy": 5.595451688766479,
      "epoch": 0.42715425014588604,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.3896,
      "mean_token_accuracy": 0.16756030768156052,
      "num_tokens": 9520490.0,
      "step": 5490
    },
    {
      "entropy": 5.626486682891846,
      "epoch": 0.42754327951760357,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.4421,
      "mean_token_accuracy": 0.16567257046699524,
      "num_tokens": 9528474.0,
      "step": 5495
    },
    {
      "entropy": 5.688561153411865,
      "epoch": 0.42793230888932116,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16336488127708435,
      "num_tokens": 9537747.0,
      "step": 5500
    },
    {
      "entropy": 5.632611274719238,
      "epoch": 0.4283213382610387,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.4534,
      "mean_token_accuracy": 0.16207062304019929,
      "num_tokens": 9546414.0,
      "step": 5505
    },
    {
      "entropy": 5.664350414276123,
      "epoch": 0.4287103676327563,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.4661,
      "mean_token_accuracy": 0.1694653421640396,
      "num_tokens": 9555166.0,
      "step": 5510
    },
    {
      "entropy": 5.688541412353516,
      "epoch": 0.42909939700447386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.519,
      "mean_token_accuracy": 0.16417355090379715,
      "num_tokens": 9563217.0,
      "step": 5515
    },
    {
      "entropy": 5.645942831039429,
      "epoch": 0.4294884263761914,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498607062195433,
      "loss": 5.5671,
      "mean_token_accuracy": 0.15941138565540314,
      "num_tokens": 9572837.0,
      "step": 5520
    },
    {
      "entropy": 5.718004369735718,
      "epoch": 0.429877455747909,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498603981280632,
      "loss": 5.462,
      "mean_token_accuracy": 0.16318256556987762,
      "num_tokens": 9581297.0,
      "step": 5525
    },
    {
      "entropy": 5.6322996616363525,
      "epoch": 0.4302664851196265,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.5443,
      "mean_token_accuracy": 0.1617195874452591,
      "num_tokens": 9590431.0,
      "step": 5530
    },
    {
      "entropy": 5.723177671432495,
      "epoch": 0.4306555144913441,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.5917,
      "mean_token_accuracy": 0.16330402940511704,
      "num_tokens": 9598719.0,
      "step": 5535
    },
    {
      "entropy": 5.7271167755126955,
      "epoch": 0.4310445438630617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498594718179409,
      "loss": 5.5861,
      "mean_token_accuracy": 0.15970134884119033,
      "num_tokens": 9606935.0,
      "step": 5540
    },
    {
      "entropy": 5.710335636138916,
      "epoch": 0.4314335732347792,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.5288,
      "mean_token_accuracy": 0.17187016010284423,
      "num_tokens": 9615104.0,
      "step": 5545
    },
    {
      "entropy": 5.803662633895874,
      "epoch": 0.4318226026064968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.5578,
      "mean_token_accuracy": 0.15836212933063507,
      "num_tokens": 9623332.0,
      "step": 5550
    },
    {
      "entropy": 5.581446504592895,
      "epoch": 0.43221163197821433,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.3979,
      "mean_token_accuracy": 0.16912583708763124,
      "num_tokens": 9631874.0,
      "step": 5555
    },
    {
      "entropy": 5.609959077835083,
      "epoch": 0.4326006613499319,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.4509,
      "mean_token_accuracy": 0.1646541029214859,
      "num_tokens": 9641150.0,
      "step": 5560
    },
    {
      "entropy": 5.571775960922241,
      "epoch": 0.4329896907216495,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.2773,
      "mean_token_accuracy": 0.1755918726325035,
      "num_tokens": 9649116.0,
      "step": 5565
    },
    {
      "entropy": 5.6169108867645265,
      "epoch": 0.43337872009336703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.5178,
      "mean_token_accuracy": 0.166967211663723,
      "num_tokens": 9657579.0,
      "step": 5570
    },
    {
      "entropy": 5.635841417312622,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.4908,
      "mean_token_accuracy": 0.1659495487809181,
      "num_tokens": 9666242.0,
      "step": 5575
    },
    {
      "entropy": 5.669478797912598,
      "epoch": 0.4341567788368022,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498569867301064,
      "loss": 5.4663,
      "mean_token_accuracy": 0.16676506102085115,
      "num_tokens": 9675196.0,
      "step": 5580
    },
    {
      "entropy": 5.643025684356689,
      "epoch": 0.43454580820851973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.4287,
      "mean_token_accuracy": 0.16679195463657379,
      "num_tokens": 9683681.0,
      "step": 5585
    },
    {
      "entropy": 5.592366361618042,
      "epoch": 0.4349348375802373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1720845341682434,
      "num_tokens": 9692445.0,
      "step": 5590
    },
    {
      "entropy": 5.6447841167449955,
      "epoch": 0.43532386695195485,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.3954,
      "mean_token_accuracy": 0.1695060133934021,
      "num_tokens": 9701256.0,
      "step": 5595
    },
    {
      "entropy": 5.647923994064331,
      "epoch": 0.43571289632367244,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16670031696558,
      "num_tokens": 9710351.0,
      "step": 5600
    },
    {
      "entropy": 5.681913232803344,
      "epoch": 0.43610192569539,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.5098,
      "mean_token_accuracy": 0.15940970182418823,
      "num_tokens": 9718795.0,
      "step": 5605
    },
    {
      "entropy": 5.679136896133423,
      "epoch": 0.43649095506710756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.4323,
      "mean_token_accuracy": 0.16713555306196212,
      "num_tokens": 9727136.0,
      "step": 5610
    },
    {
      "entropy": 5.578325414657593,
      "epoch": 0.43687998443882514,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.4182,
      "mean_token_accuracy": 0.17000139653682708,
      "num_tokens": 9735712.0,
      "step": 5615
    },
    {
      "entropy": 5.6629225730896,
      "epoch": 0.4372690138105427,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.4542,
      "mean_token_accuracy": 0.16782625317573546,
      "num_tokens": 9744027.0,
      "step": 5620
    },
    {
      "entropy": 5.750279188156128,
      "epoch": 0.43765804318226026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.59,
      "mean_token_accuracy": 0.1612232133746147,
      "num_tokens": 9753083.0,
      "step": 5625
    },
    {
      "entropy": 5.72616057395935,
      "epoch": 0.43804707255397785,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.5179,
      "mean_token_accuracy": 0.16758409440517424,
      "num_tokens": 9762098.0,
      "step": 5630
    },
    {
      "entropy": 5.692498731613159,
      "epoch": 0.4384361019256954,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.4615,
      "mean_token_accuracy": 0.16436221897602082,
      "num_tokens": 9770956.0,
      "step": 5635
    },
    {
      "entropy": 5.574732780456543,
      "epoch": 0.43882513129741296,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1726594388484955,
      "num_tokens": 9779427.0,
      "step": 5640
    },
    {
      "entropy": 5.621200513839722,
      "epoch": 0.4392141606691305,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.4777,
      "mean_token_accuracy": 0.17316209226846696,
      "num_tokens": 9787655.0,
      "step": 5645
    },
    {
      "entropy": 5.668504667282105,
      "epoch": 0.4396031900408481,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.4158,
      "mean_token_accuracy": 0.1715027719736099,
      "num_tokens": 9796850.0,
      "step": 5650
    },
    {
      "entropy": 5.651020812988281,
      "epoch": 0.43999221941256567,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.5034,
      "mean_token_accuracy": 0.17284105122089385,
      "num_tokens": 9805317.0,
      "step": 5655
    },
    {
      "entropy": 5.650268125534057,
      "epoch": 0.4403812487842832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.4416,
      "mean_token_accuracy": 0.17098015695810317,
      "num_tokens": 9814249.0,
      "step": 5660
    },
    {
      "entropy": 5.67411699295044,
      "epoch": 0.4407702781560008,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.539,
      "mean_token_accuracy": 0.16412654519081116,
      "num_tokens": 9823122.0,
      "step": 5665
    },
    {
      "entropy": 5.595461463928222,
      "epoch": 0.44115930752771837,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.4086,
      "mean_token_accuracy": 0.17494799494743346,
      "num_tokens": 9831470.0,
      "step": 5670
    },
    {
      "entropy": 5.623825263977051,
      "epoch": 0.4415483368994359,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.5128,
      "mean_token_accuracy": 0.16640195101499558,
      "num_tokens": 9840096.0,
      "step": 5675
    },
    {
      "entropy": 5.687500905990601,
      "epoch": 0.4419373662711535,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.4607,
      "mean_token_accuracy": 0.17098743170499803,
      "num_tokens": 9848595.0,
      "step": 5680
    },
    {
      "entropy": 5.637724256515503,
      "epoch": 0.442326395642871,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.4073,
      "mean_token_accuracy": 0.16370895206928254,
      "num_tokens": 9857353.0,
      "step": 5685
    },
    {
      "entropy": 5.63226637840271,
      "epoch": 0.4427154250145886,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.46,
      "mean_token_accuracy": 0.16471842527389527,
      "num_tokens": 9865828.0,
      "step": 5690
    },
    {
      "entropy": 5.593899774551391,
      "epoch": 0.4431044543863062,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.4735,
      "mean_token_accuracy": 0.16496108174324037,
      "num_tokens": 9874540.0,
      "step": 5695
    },
    {
      "entropy": 5.677505302429199,
      "epoch": 0.4434934837580237,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.4622,
      "mean_token_accuracy": 0.16470122337341309,
      "num_tokens": 9883046.0,
      "step": 5700
    },
    {
      "entropy": 5.6196901321411135,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049849080908704,
      "loss": 5.4166,
      "mean_token_accuracy": 0.16411941945552827,
      "num_tokens": 9892282.0,
      "step": 5705
    },
    {
      "entropy": 5.635061359405517,
      "epoch": 0.44427154250145884,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.3502,
      "mean_token_accuracy": 0.17350826859474183,
      "num_tokens": 9900653.0,
      "step": 5710
    },
    {
      "entropy": 5.568204784393311,
      "epoch": 0.4446605718731764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498484392859752,
      "loss": 5.3852,
      "mean_token_accuracy": 0.17214204668998717,
      "num_tokens": 9909323.0,
      "step": 5715
    },
    {
      "entropy": 5.582053184509277,
      "epoch": 0.445049601244894,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.5187,
      "mean_token_accuracy": 0.16376986354589462,
      "num_tokens": 9918155.0,
      "step": 5720
    },
    {
      "entropy": 5.747524881362915,
      "epoch": 0.44543863061661154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.4763,
      "mean_token_accuracy": 0.16589940339326859,
      "num_tokens": 9925977.0,
      "step": 5725
    },
    {
      "entropy": 5.637293529510498,
      "epoch": 0.44582765998832913,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.4612,
      "mean_token_accuracy": 0.16873055398464204,
      "num_tokens": 9934606.0,
      "step": 5730
    },
    {
      "entropy": 5.580166482925415,
      "epoch": 0.44621668936004666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.4339,
      "mean_token_accuracy": 0.16639334559440613,
      "num_tokens": 9943638.0,
      "step": 5735
    },
    {
      "entropy": 5.641740894317627,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.4527,
      "mean_token_accuracy": 0.16659653931856155,
      "num_tokens": 9952686.0,
      "step": 5740
    },
    {
      "entropy": 5.693140602111816,
      "epoch": 0.44699474810348183,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.6415,
      "mean_token_accuracy": 0.15615570396184922,
      "num_tokens": 9961947.0,
      "step": 5745
    },
    {
      "entropy": 5.670505428314209,
      "epoch": 0.44738377747519936,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.4647,
      "mean_token_accuracy": 0.16673058420419692,
      "num_tokens": 9971218.0,
      "step": 5750
    },
    {
      "entropy": 5.664007711410522,
      "epoch": 0.44777280684691695,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1662154719233513,
      "num_tokens": 9980741.0,
      "step": 5755
    },
    {
      "entropy": 5.694485855102539,
      "epoch": 0.4481618362186345,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498455351987111,
      "loss": 5.4705,
      "mean_token_accuracy": 0.16652709394693374,
      "num_tokens": 9989483.0,
      "step": 5760
    },
    {
      "entropy": 5.693090534210205,
      "epoch": 0.44855086559035207,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16687163412570954,
      "num_tokens": 9998491.0,
      "step": 5765
    },
    {
      "entropy": 5.543629884719849,
      "epoch": 0.44893989496206965,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.3806,
      "mean_token_accuracy": 0.17117684185504914,
      "num_tokens": 10006276.0,
      "step": 5770
    },
    {
      "entropy": 5.5802089214324955,
      "epoch": 0.4493289243337872,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498445610664037,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16785627007484435,
      "num_tokens": 10015088.0,
      "step": 5775
    },
    {
      "entropy": 5.663906669616699,
      "epoch": 0.44971795370550477,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498442356775417,
      "loss": 5.4266,
      "mean_token_accuracy": 0.16554765850305558,
      "num_tokens": 10023305.0,
      "step": 5780
    },
    {
      "entropy": 5.575909042358399,
      "epoch": 0.45010698307722236,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.3805,
      "mean_token_accuracy": 0.16881942003965378,
      "num_tokens": 10031540.0,
      "step": 5785
    },
    {
      "entropy": 5.6279661655426025,
      "epoch": 0.4504960124489399,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049843583882708,
      "loss": 5.4733,
      "mean_token_accuracy": 0.16908168643712998,
      "num_tokens": 10040025.0,
      "step": 5790
    },
    {
      "entropy": 5.6401628971099855,
      "epoch": 0.4508850418206575,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498432574767462,
      "loss": 5.48,
      "mean_token_accuracy": 0.16680154353380203,
      "num_tokens": 10048904.0,
      "step": 5795
    },
    {
      "entropy": 5.644533729553222,
      "epoch": 0.451274071192375,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.5133,
      "mean_token_accuracy": 0.16642902791500092,
      "num_tokens": 10057608.0,
      "step": 5800
    },
    {
      "entropy": 5.713375568389893,
      "epoch": 0.4516631005640926,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.4701,
      "mean_token_accuracy": 0.17314359098672866,
      "num_tokens": 10066587.0,
      "step": 5805
    },
    {
      "entropy": 5.608742380142212,
      "epoch": 0.4520521299358102,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498422762247404,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1667902648448944,
      "num_tokens": 10075651.0,
      "step": 5810
    },
    {
      "entropy": 5.65361647605896,
      "epoch": 0.4524411593075277,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.5379,
      "mean_token_accuracy": 0.16456001549959182,
      "num_tokens": 10084158.0,
      "step": 5815
    },
    {
      "entropy": 5.714414310455322,
      "epoch": 0.4528301886792453,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.4715,
      "mean_token_accuracy": 0.16263072341680526,
      "num_tokens": 10093730.0,
      "step": 5820
    },
    {
      "entropy": 5.636698055267334,
      "epoch": 0.4532192180509628,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.3956,
      "mean_token_accuracy": 0.16606880724430084,
      "num_tokens": 10102457.0,
      "step": 5825
    },
    {
      "entropy": 5.657480669021607,
      "epoch": 0.4536082474226804,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.5361,
      "mean_token_accuracy": 0.1659406006336212,
      "num_tokens": 10111188.0,
      "step": 5830
    },
    {
      "entropy": 5.691649389266968,
      "epoch": 0.453997276794398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.4321,
      "mean_token_accuracy": 0.1693912699818611,
      "num_tokens": 10120268.0,
      "step": 5835
    },
    {
      "entropy": 5.646623325347901,
      "epoch": 0.45438630616611553,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.5018,
      "mean_token_accuracy": 0.159344282746315,
      "num_tokens": 10129262.0,
      "step": 5840
    },
    {
      "entropy": 5.604523992538452,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.4425,
      "mean_token_accuracy": 0.17042097300291062,
      "num_tokens": 10137417.0,
      "step": 5845
    },
    {
      "entropy": 5.60966420173645,
      "epoch": 0.45516436490955064,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.3859,
      "mean_token_accuracy": 0.169807331264019,
      "num_tokens": 10145804.0,
      "step": 5850
    },
    {
      "entropy": 5.6830277919769285,
      "epoch": 0.45555339428126823,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.5243,
      "mean_token_accuracy": 0.1644147291779518,
      "num_tokens": 10154421.0,
      "step": 5855
    },
    {
      "entropy": 5.750033950805664,
      "epoch": 0.4559424236529858,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.5348,
      "mean_token_accuracy": 0.16695573478937148,
      "num_tokens": 10163631.0,
      "step": 5860
    },
    {
      "entropy": 5.64238224029541,
      "epoch": 0.45633145302470335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1701275646686554,
      "num_tokens": 10172926.0,
      "step": 5865
    },
    {
      "entropy": 5.679831504821777,
      "epoch": 0.45672048239642093,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.5529,
      "mean_token_accuracy": 0.15790846645832063,
      "num_tokens": 10181671.0,
      "step": 5870
    },
    {
      "entropy": 5.6128734111785885,
      "epoch": 0.4571095117681385,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.3875,
      "mean_token_accuracy": 0.17633408457040786,
      "num_tokens": 10189626.0,
      "step": 5875
    },
    {
      "entropy": 5.681553602218628,
      "epoch": 0.45749854113985605,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.5232,
      "mean_token_accuracy": 0.16761491298675538,
      "num_tokens": 10198212.0,
      "step": 5880
    },
    {
      "entropy": 5.586812591552734,
      "epoch": 0.45788757051157364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.4185,
      "mean_token_accuracy": 0.16471427530050278,
      "num_tokens": 10207342.0,
      "step": 5885
    },
    {
      "entropy": 5.704931020736694,
      "epoch": 0.45827659988329117,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.5231,
      "mean_token_accuracy": 0.15820106565952302,
      "num_tokens": 10216413.0,
      "step": 5890
    },
    {
      "entropy": 5.657903099060059,
      "epoch": 0.45866562925500876,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.4524,
      "mean_token_accuracy": 0.16937243938446045,
      "num_tokens": 10225889.0,
      "step": 5895
    },
    {
      "entropy": 5.589326333999634,
      "epoch": 0.45905465862672634,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.4063,
      "mean_token_accuracy": 0.17112260311841965,
      "num_tokens": 10234518.0,
      "step": 5900
    },
    {
      "entropy": 5.695744514465332,
      "epoch": 0.4594436879984439,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.5391,
      "mean_token_accuracy": 0.16143682301044465,
      "num_tokens": 10243074.0,
      "step": 5905
    },
    {
      "entropy": 5.771560573577881,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.4887,
      "mean_token_accuracy": 0.16415552347898482,
      "num_tokens": 10253473.0,
      "step": 5910
    },
    {
      "entropy": 5.69731650352478,
      "epoch": 0.460221746741879,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1648534968495369,
      "num_tokens": 10261866.0,
      "step": 5915
    },
    {
      "entropy": 5.649386692047119,
      "epoch": 0.4606107761135966,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1710118681192398,
      "num_tokens": 10270350.0,
      "step": 5920
    },
    {
      "entropy": 5.647009754180909,
      "epoch": 0.46099980548531416,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.5711,
      "mean_token_accuracy": 0.15795120000839233,
      "num_tokens": 10279556.0,
      "step": 5925
    },
    {
      "entropy": 5.586729669570923,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1722436711192131,
      "num_tokens": 10287489.0,
      "step": 5930
    },
    {
      "entropy": 5.604912328720093,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.3242,
      "mean_token_accuracy": 0.16806061565876007,
      "num_tokens": 10296280.0,
      "step": 5935
    },
    {
      "entropy": 5.619849300384521,
      "epoch": 0.4621668936004668,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1619819775223732,
      "num_tokens": 10304091.0,
      "step": 5940
    },
    {
      "entropy": 5.668075275421143,
      "epoch": 0.4625559229721844,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.5026,
      "mean_token_accuracy": 0.15887704938650132,
      "num_tokens": 10313058.0,
      "step": 5945
    },
    {
      "entropy": 5.622173261642456,
      "epoch": 0.462944952343902,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.4749,
      "mean_token_accuracy": 0.1640453204512596,
      "num_tokens": 10321642.0,
      "step": 5950
    },
    {
      "entropy": 5.577755784988403,
      "epoch": 0.4633339817156195,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.2922,
      "mean_token_accuracy": 0.17254432886838914,
      "num_tokens": 10329829.0,
      "step": 5955
    },
    {
      "entropy": 5.640766525268555,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.5284,
      "mean_token_accuracy": 0.16286519914865494,
      "num_tokens": 10338182.0,
      "step": 5960
    },
    {
      "entropy": 5.673437929153442,
      "epoch": 0.46411204045905463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.4208,
      "mean_token_accuracy": 0.1703182965517044,
      "num_tokens": 10346377.0,
      "step": 5965
    },
    {
      "entropy": 5.668019008636475,
      "epoch": 0.4645010698307722,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.4089,
      "mean_token_accuracy": 0.17171140760183334,
      "num_tokens": 10354937.0,
      "step": 5970
    },
    {
      "entropy": 5.543028974533081,
      "epoch": 0.4648900992024898,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.3114,
      "mean_token_accuracy": 0.17455710768699645,
      "num_tokens": 10363706.0,
      "step": 5975
    },
    {
      "entropy": 5.574260568618774,
      "epoch": 0.46527912857420733,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16650742590427398,
      "num_tokens": 10372292.0,
      "step": 5980
    },
    {
      "entropy": 5.671193408966064,
      "epoch": 0.4656681579459249,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.4386,
      "mean_token_accuracy": 0.17165132462978364,
      "num_tokens": 10381260.0,
      "step": 5985
    },
    {
      "entropy": 5.648577165603638,
      "epoch": 0.4660571873176425,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16947742700576782,
      "num_tokens": 10390646.0,
      "step": 5990
    },
    {
      "entropy": 5.625260925292968,
      "epoch": 0.46644621668936004,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16828270852565766,
      "num_tokens": 10399056.0,
      "step": 5995
    },
    {
      "entropy": 5.6372960090637205,
      "epoch": 0.4668352460610776,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.5048,
      "mean_token_accuracy": 0.16254437565803528,
      "num_tokens": 10408746.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.468773723818914,
      "eval_loss": 5.478248119354248,
      "eval_mean_token_accuracy": 0.17529629864178678,
      "eval_num_tokens": 10408746.0,
      "eval_runtime": 21.786,
      "eval_samples_per_second": 1907.559,
      "eval_steps_per_second": 238.456,
      "step": 6000
    },
    {
      "entropy": 5.667443513870239,
      "epoch": 0.46722427543279516,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498292423500673,
      "loss": 5.4276,
      "mean_token_accuracy": 0.16848916113376616,
      "num_tokens": 10416937.0,
      "step": 6005
    },
    {
      "entropy": 5.6209752559661865,
      "epoch": 0.46761330480451274,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004982890137064102,
      "loss": 5.4023,
      "mean_token_accuracy": 0.17630034387111665,
      "num_tokens": 10426343.0,
      "step": 6010
    },
    {
      "entropy": 5.6079370975494385,
      "epoch": 0.46800233417623033,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498285600524092,
      "loss": 5.4317,
      "mean_token_accuracy": 0.1709451973438263,
      "num_tokens": 10435014.0,
      "step": 6015
    },
    {
      "entropy": 5.557825231552124,
      "epoch": 0.46839136354794786,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982821839537701,
      "loss": 5.447,
      "mean_token_accuracy": 0.16832986921072007,
      "num_tokens": 10443522.0,
      "step": 6020
    },
    {
      "entropy": 5.6745398998260494,
      "epoch": 0.46878039291966545,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982787639954966,
      "loss": 5.4939,
      "mean_token_accuracy": 0.1648608475923538,
      "num_tokens": 10452202.0,
      "step": 6025
    },
    {
      "entropy": 5.631177759170532,
      "epoch": 0.469169422291383,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982753406493232,
      "loss": 5.4735,
      "mean_token_accuracy": 0.16007108837366105,
      "num_tokens": 10462293.0,
      "step": 6030
    },
    {
      "entropy": 5.586612892150879,
      "epoch": 0.46955845166310056,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982719139153018,
      "loss": 5.4051,
      "mean_token_accuracy": 0.1679594933986664,
      "num_tokens": 10470322.0,
      "step": 6035
    },
    {
      "entropy": 5.635045433044434,
      "epoch": 0.46994748103481815,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982684837934845,
      "loss": 5.3935,
      "mean_token_accuracy": 0.17465801239013673,
      "num_tokens": 10479581.0,
      "step": 6040
    },
    {
      "entropy": 5.655844640731812,
      "epoch": 0.4703365104065357,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982650502839234,
      "loss": 5.4006,
      "mean_token_accuracy": 0.1719515100121498,
      "num_tokens": 10488207.0,
      "step": 6045
    },
    {
      "entropy": 5.579202795028687,
      "epoch": 0.47072553977825327,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982616133866705,
      "loss": 5.4117,
      "mean_token_accuracy": 0.16743631064891815,
      "num_tokens": 10497360.0,
      "step": 6050
    },
    {
      "entropy": 5.578354835510254,
      "epoch": 0.4711145691499708,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982581731017779,
      "loss": 5.3959,
      "mean_token_accuracy": 0.16868714839220048,
      "num_tokens": 10506131.0,
      "step": 6055
    },
    {
      "entropy": 5.615327072143555,
      "epoch": 0.4715035985216884,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982547294292979,
      "loss": 5.4404,
      "mean_token_accuracy": 0.16701407134532928,
      "num_tokens": 10514955.0,
      "step": 6060
    },
    {
      "entropy": 5.593543672561646,
      "epoch": 0.47189262789340597,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982512823692828,
      "loss": 5.4119,
      "mean_token_accuracy": 0.16133114844560623,
      "num_tokens": 10523743.0,
      "step": 6065
    },
    {
      "entropy": 5.605992555618286,
      "epoch": 0.4722816572651235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982478319217848,
      "loss": 5.3815,
      "mean_token_accuracy": 0.1775616154074669,
      "num_tokens": 10533149.0,
      "step": 6070
    },
    {
      "entropy": 5.565766048431397,
      "epoch": 0.4726706866368411,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982443780868562,
      "loss": 5.417,
      "mean_token_accuracy": 0.17523194700479508,
      "num_tokens": 10541291.0,
      "step": 6075
    },
    {
      "entropy": 5.651870632171631,
      "epoch": 0.4730597160085587,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982409208645496,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16842615902423858,
      "num_tokens": 10550005.0,
      "step": 6080
    },
    {
      "entropy": 5.692563199996949,
      "epoch": 0.4734487453802762,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982374602549173,
      "loss": 5.4792,
      "mean_token_accuracy": 0.16938978135585786,
      "num_tokens": 10558957.0,
      "step": 6085
    },
    {
      "entropy": 5.64795618057251,
      "epoch": 0.4738377747519938,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982339962580119,
      "loss": 5.3678,
      "mean_token_accuracy": 0.16741629391908647,
      "num_tokens": 10567293.0,
      "step": 6090
    },
    {
      "entropy": 5.580725526809692,
      "epoch": 0.4742268041237113,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982305288738859,
      "loss": 5.3384,
      "mean_token_accuracy": 0.17617610692977906,
      "num_tokens": 10575546.0,
      "step": 6095
    },
    {
      "entropy": 5.528723192214966,
      "epoch": 0.4746158334954289,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004982270581025919,
      "loss": 5.465,
      "mean_token_accuracy": 0.17209935039281846,
      "num_tokens": 10584779.0,
      "step": 6100
    },
    {
      "entropy": 5.587629985809326,
      "epoch": 0.4750048628671465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982235839441826,
      "loss": 5.3955,
      "mean_token_accuracy": 0.17494287937879563,
      "num_tokens": 10592852.0,
      "step": 6105
    },
    {
      "entropy": 5.655500841140747,
      "epoch": 0.475393892238864,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982201063987108,
      "loss": 5.4999,
      "mean_token_accuracy": 0.1644456207752228,
      "num_tokens": 10601875.0,
      "step": 6110
    },
    {
      "entropy": 5.611720180511474,
      "epoch": 0.4757829216105816,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982166254662292,
      "loss": 5.4688,
      "mean_token_accuracy": 0.17126654237508773,
      "num_tokens": 10610438.0,
      "step": 6115
    },
    {
      "entropy": 5.627661180496216,
      "epoch": 0.47617195098229914,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982131411467904,
      "loss": 5.4636,
      "mean_token_accuracy": 0.16586989760398865,
      "num_tokens": 10619479.0,
      "step": 6120
    },
    {
      "entropy": 5.612498712539673,
      "epoch": 0.4765609803540167,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982096534404476,
      "loss": 5.3893,
      "mean_token_accuracy": 0.17240849733352662,
      "num_tokens": 10628519.0,
      "step": 6125
    },
    {
      "entropy": 5.640328454971313,
      "epoch": 0.4769500097257343,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982061623472535,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17426417917013168,
      "num_tokens": 10636522.0,
      "step": 6130
    },
    {
      "entropy": 5.569852495193482,
      "epoch": 0.47733903909745184,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982026678672612,
      "loss": 5.3648,
      "mean_token_accuracy": 0.16899030655622482,
      "num_tokens": 10645728.0,
      "step": 6135
    },
    {
      "entropy": 5.596399021148682,
      "epoch": 0.47772806846916943,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981991700005238,
      "loss": 5.4789,
      "mean_token_accuracy": 0.1673264682292938,
      "num_tokens": 10654868.0,
      "step": 6140
    },
    {
      "entropy": 5.548848628997803,
      "epoch": 0.47811709784088696,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498195668747094,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17227984219789505,
      "num_tokens": 10662958.0,
      "step": 6145
    },
    {
      "entropy": 5.6338269233703615,
      "epoch": 0.47850612721260455,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981921641070254,
      "loss": 5.3796,
      "mean_token_accuracy": 0.17123519778251647,
      "num_tokens": 10671333.0,
      "step": 6150
    },
    {
      "entropy": 5.627652263641357,
      "epoch": 0.47889515658432213,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981886560803708,
      "loss": 5.4014,
      "mean_token_accuracy": 0.1699555292725563,
      "num_tokens": 10680523.0,
      "step": 6155
    },
    {
      "entropy": 5.619808626174927,
      "epoch": 0.47928418595603967,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981851446671838,
      "loss": 5.4687,
      "mean_token_accuracy": 0.16397960782051085,
      "num_tokens": 10689656.0,
      "step": 6160
    },
    {
      "entropy": 5.666473770141602,
      "epoch": 0.47967321532775725,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004981816298675173,
      "loss": 5.5385,
      "mean_token_accuracy": 0.16096996814012526,
      "num_tokens": 10698423.0,
      "step": 6165
    },
    {
      "entropy": 5.625974845886231,
      "epoch": 0.4800622446994748,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981781116814248,
      "loss": 5.4071,
      "mean_token_accuracy": 0.1723208859562874,
      "num_tokens": 10706760.0,
      "step": 6170
    },
    {
      "entropy": 5.707794189453125,
      "epoch": 0.48045127407119237,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981745901089596,
      "loss": 5.5088,
      "mean_token_accuracy": 0.1589950516819954,
      "num_tokens": 10716221.0,
      "step": 6175
    },
    {
      "entropy": 5.5495470523834225,
      "epoch": 0.48084030344290996,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981710651501753,
      "loss": 5.3102,
      "mean_token_accuracy": 0.18165120333433152,
      "num_tokens": 10724603.0,
      "step": 6180
    },
    {
      "entropy": 5.645468711853027,
      "epoch": 0.4812293328146275,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981675368051254,
      "loss": 5.5351,
      "mean_token_accuracy": 0.15899534821510314,
      "num_tokens": 10733577.0,
      "step": 6185
    },
    {
      "entropy": 5.712690258026123,
      "epoch": 0.4816183621863451,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004981640050738633,
      "loss": 5.4466,
      "mean_token_accuracy": 0.17270689308643342,
      "num_tokens": 10743199.0,
      "step": 6190
    },
    {
      "entropy": 5.616707849502563,
      "epoch": 0.48200739155806266,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981604699564426,
      "loss": 5.3312,
      "mean_token_accuracy": 0.17357029765844345,
      "num_tokens": 10751734.0,
      "step": 6195
    },
    {
      "entropy": 5.6283769607543945,
      "epoch": 0.4823964209297802,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004981569314529169,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16709917336702346,
      "num_tokens": 10760194.0,
      "step": 6200
    },
    {
      "entropy": 5.650402545928955,
      "epoch": 0.4827854503014978,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049815338956334,
      "loss": 5.4036,
      "mean_token_accuracy": 0.17521733343601226,
      "num_tokens": 10768944.0,
      "step": 6205
    },
    {
      "entropy": 5.69504828453064,
      "epoch": 0.4831744796732153,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004981498442877656,
      "loss": 5.5049,
      "mean_token_accuracy": 0.1730547532439232,
      "num_tokens": 10777685.0,
      "step": 6210
    },
    {
      "entropy": 5.642111921310425,
      "epoch": 0.4835635090449329,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981462956262474,
      "loss": 5.4765,
      "mean_token_accuracy": 0.16693192422389985,
      "num_tokens": 10787264.0,
      "step": 6215
    },
    {
      "entropy": 5.602992391586303,
      "epoch": 0.4839525384166505,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981427435788396,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1655664786696434,
      "num_tokens": 10796555.0,
      "step": 6220
    },
    {
      "entropy": 5.525999927520752,
      "epoch": 0.484341567788368,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981391881455957,
      "loss": 5.3515,
      "mean_token_accuracy": 0.17589380443096161,
      "num_tokens": 10805379.0,
      "step": 6225
    },
    {
      "entropy": 5.565594863891602,
      "epoch": 0.4847305971600856,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981356293265696,
      "loss": 5.4786,
      "mean_token_accuracy": 0.16520614624023439,
      "num_tokens": 10815406.0,
      "step": 6230
    },
    {
      "entropy": 5.658706951141357,
      "epoch": 0.4851196265318031,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981320671218157,
      "loss": 5.5369,
      "mean_token_accuracy": 0.16370739042758942,
      "num_tokens": 10824443.0,
      "step": 6235
    },
    {
      "entropy": 5.686445903778076,
      "epoch": 0.4855086559035207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981285015313877,
      "loss": 5.366,
      "mean_token_accuracy": 0.1736448109149933,
      "num_tokens": 10833497.0,
      "step": 6240
    },
    {
      "entropy": 5.557299995422364,
      "epoch": 0.4858976852752383,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981249325553399,
      "loss": 5.4135,
      "mean_token_accuracy": 0.1717936083674431,
      "num_tokens": 10841864.0,
      "step": 6245
    },
    {
      "entropy": 5.662262201309204,
      "epoch": 0.48628671464695583,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981213601937264,
      "loss": 5.4643,
      "mean_token_accuracy": 0.17284016013145448,
      "num_tokens": 10850887.0,
      "step": 6250
    },
    {
      "entropy": 5.590152788162231,
      "epoch": 0.4866757440186734,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981177844466013,
      "loss": 5.3453,
      "mean_token_accuracy": 0.174080491065979,
      "num_tokens": 10859629.0,
      "step": 6255
    },
    {
      "entropy": 5.652895593643189,
      "epoch": 0.48706477339039095,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981142053140192,
      "loss": 5.5285,
      "mean_token_accuracy": 0.16324191689491271,
      "num_tokens": 10868457.0,
      "step": 6260
    },
    {
      "entropy": 5.6257092475891115,
      "epoch": 0.48745380276210853,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981106227960339,
      "loss": 5.4483,
      "mean_token_accuracy": 0.16676026433706284,
      "num_tokens": 10878007.0,
      "step": 6265
    },
    {
      "entropy": 5.703521251678467,
      "epoch": 0.4878428321338261,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981070368927001,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16467054486274718,
      "num_tokens": 10887689.0,
      "step": 6270
    },
    {
      "entropy": 5.585311794281006,
      "epoch": 0.48823186150554365,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498103447604072,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17395345568656922,
      "num_tokens": 10896214.0,
      "step": 6275
    },
    {
      "entropy": 5.55946979522705,
      "epoch": 0.48862089087726124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980998549302044,
      "loss": 5.4149,
      "mean_token_accuracy": 0.16991112530231475,
      "num_tokens": 10904946.0,
      "step": 6280
    },
    {
      "entropy": 5.693157911300659,
      "epoch": 0.4890099202489788,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980962588711516,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1608987919986248,
      "num_tokens": 10913382.0,
      "step": 6285
    },
    {
      "entropy": 5.672442865371704,
      "epoch": 0.48939894962069636,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498092659426968,
      "loss": 5.4338,
      "mean_token_accuracy": 0.17338544130325317,
      "num_tokens": 10922055.0,
      "step": 6290
    },
    {
      "entropy": 5.615342998504639,
      "epoch": 0.48978797899241394,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980890565977085,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16947162002325059,
      "num_tokens": 10930282.0,
      "step": 6295
    },
    {
      "entropy": 5.565211153030395,
      "epoch": 0.4901770083641315,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980854503834276,
      "loss": 5.4367,
      "mean_token_accuracy": 0.17593979835510254,
      "num_tokens": 10938432.0,
      "step": 6300
    },
    {
      "entropy": 5.670564031600952,
      "epoch": 0.49056603773584906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980818407841802,
      "loss": 5.4375,
      "mean_token_accuracy": 0.17329999804496765,
      "num_tokens": 10946354.0,
      "step": 6305
    },
    {
      "entropy": 5.725423860549927,
      "epoch": 0.49095506710756665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980782278000207,
      "loss": 5.4683,
      "mean_token_accuracy": 0.16744253635406495,
      "num_tokens": 10954885.0,
      "step": 6310
    },
    {
      "entropy": 5.590633678436279,
      "epoch": 0.4913440964792842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980746114310043,
      "loss": 5.3743,
      "mean_token_accuracy": 0.1690499573945999,
      "num_tokens": 10963173.0,
      "step": 6315
    },
    {
      "entropy": 5.634382104873657,
      "epoch": 0.49173312585100176,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980709916771858,
      "loss": 5.5258,
      "mean_token_accuracy": 0.15521946102380751,
      "num_tokens": 10972264.0,
      "step": 6320
    },
    {
      "entropy": 5.617056846618652,
      "epoch": 0.4921221552227193,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980673685386198,
      "loss": 5.3702,
      "mean_token_accuracy": 0.1731475278735161,
      "num_tokens": 10980645.0,
      "step": 6325
    },
    {
      "entropy": 5.541935157775879,
      "epoch": 0.4925111845944369,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980637420153618,
      "loss": 5.33,
      "mean_token_accuracy": 0.17462508678436278,
      "num_tokens": 10989121.0,
      "step": 6330
    },
    {
      "entropy": 5.581294679641724,
      "epoch": 0.49290021396615447,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980601121074664,
      "loss": 5.3333,
      "mean_token_accuracy": 0.18154052793979644,
      "num_tokens": 10998249.0,
      "step": 6335
    },
    {
      "entropy": 5.693126916885376,
      "epoch": 0.493289243337872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980564788149889,
      "loss": 5.448,
      "mean_token_accuracy": 0.16501189395785332,
      "num_tokens": 11007470.0,
      "step": 6340
    },
    {
      "entropy": 5.657134103775024,
      "epoch": 0.4936782727095896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980528421379844,
      "loss": 5.3946,
      "mean_token_accuracy": 0.1677241399884224,
      "num_tokens": 11015912.0,
      "step": 6345
    },
    {
      "entropy": 5.585910654067993,
      "epoch": 0.4940673020813071,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498049202076508,
      "loss": 5.418,
      "mean_token_accuracy": 0.16667619943618775,
      "num_tokens": 11024629.0,
      "step": 6350
    },
    {
      "entropy": 5.614324712753296,
      "epoch": 0.4944563314530247,
      "grad_norm": 1.0,
      "learning_rate": 0.0004980455586306149,
      "loss": 5.3509,
      "mean_token_accuracy": 0.1716369092464447,
      "num_tokens": 11033953.0,
      "step": 6355
    },
    {
      "entropy": 5.545697212219238,
      "epoch": 0.4948453608247423,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980419118003605,
      "loss": 5.4262,
      "mean_token_accuracy": 0.1693047046661377,
      "num_tokens": 11042891.0,
      "step": 6360
    },
    {
      "entropy": 5.601671361923218,
      "epoch": 0.4952343901964598,
      "grad_norm": 1.0,
      "learning_rate": 0.0004980382615858001,
      "loss": 5.5162,
      "mean_token_accuracy": 0.16135913282632827,
      "num_tokens": 11051262.0,
      "step": 6365
    },
    {
      "entropy": 5.606281852722168,
      "epoch": 0.4956234195681774,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980346079869892,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1761609584093094,
      "num_tokens": 11059357.0,
      "step": 6370
    },
    {
      "entropy": 5.603213357925415,
      "epoch": 0.49601244893989493,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498030951003983,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1736203297972679,
      "num_tokens": 11068076.0,
      "step": 6375
    },
    {
      "entropy": 5.53122091293335,
      "epoch": 0.4964014783116125,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004980272906368371,
      "loss": 5.2448,
      "mean_token_accuracy": 0.18355217725038528,
      "num_tokens": 11076804.0,
      "step": 6380
    },
    {
      "entropy": 5.650952482223511,
      "epoch": 0.4967905076833301,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980236268856071,
      "loss": 5.4407,
      "mean_token_accuracy": 0.17113036364316941,
      "num_tokens": 11085794.0,
      "step": 6385
    },
    {
      "entropy": 5.527957010269165,
      "epoch": 0.49717953705504764,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980199597503487,
      "loss": 5.3427,
      "mean_token_accuracy": 0.16856258362531662,
      "num_tokens": 11094989.0,
      "step": 6390
    },
    {
      "entropy": 5.545766878128052,
      "epoch": 0.4975685664267652,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980162892311171,
      "loss": 5.3043,
      "mean_token_accuracy": 0.18141588121652602,
      "num_tokens": 11103046.0,
      "step": 6395
    },
    {
      "entropy": 5.5342535972595215,
      "epoch": 0.4979575957984828,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980126153279684,
      "loss": 5.3777,
      "mean_token_accuracy": 0.17280755192041397,
      "num_tokens": 11111533.0,
      "step": 6400
    },
    {
      "entropy": 5.606961250305176,
      "epoch": 0.49834662517020034,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980089380409583,
      "loss": 5.5668,
      "mean_token_accuracy": 0.15929663628339769,
      "num_tokens": 11120534.0,
      "step": 6405
    },
    {
      "entropy": 5.7170703411102295,
      "epoch": 0.4987356545419179,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980052573701424,
      "loss": 5.4978,
      "mean_token_accuracy": 0.17221135199069976,
      "num_tokens": 11128354.0,
      "step": 6410
    },
    {
      "entropy": 5.644919776916504,
      "epoch": 0.49912468391363546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980015733155767,
      "loss": 5.412,
      "mean_token_accuracy": 0.16755639761686325,
      "num_tokens": 11136201.0,
      "step": 6415
    },
    {
      "entropy": 5.673887538909912,
      "epoch": 0.49951371328535304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979978858773171,
      "loss": 5.3676,
      "mean_token_accuracy": 0.1745676428079605,
      "num_tokens": 11145131.0,
      "step": 6420
    },
    {
      "entropy": 5.552023649215698,
      "epoch": 0.49990274265707063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979941950554195,
      "loss": 5.3054,
      "mean_token_accuracy": 0.1781318262219429,
      "num_tokens": 11154514.0,
      "step": 6425
    },
    {
      "entropy": 5.660540771484375,
      "epoch": 0.5002917720287882,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979905008499399,
      "loss": 5.3985,
      "mean_token_accuracy": 0.17282738238573075,
      "num_tokens": 11163160.0,
      "step": 6430
    },
    {
      "entropy": 5.576961517333984,
      "epoch": 0.5006808014005057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979868032609344,
      "loss": 5.3869,
      "mean_token_accuracy": 0.17956127524375914,
      "num_tokens": 11171573.0,
      "step": 6435
    },
    {
      "entropy": 5.56858458518982,
      "epoch": 0.5010698307722233,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979831022884591,
      "loss": 5.3739,
      "mean_token_accuracy": 0.17165141701698303,
      "num_tokens": 11180028.0,
      "step": 6440
    },
    {
      "entropy": 5.55673394203186,
      "epoch": 0.5014588601439409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979793979325701,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17160007655620574,
      "num_tokens": 11188403.0,
      "step": 6445
    },
    {
      "entropy": 5.49115891456604,
      "epoch": 0.5018478895156584,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979756901933236,
      "loss": 5.2994,
      "mean_token_accuracy": 0.18273988515138626,
      "num_tokens": 11196391.0,
      "step": 6450
    },
    {
      "entropy": 5.588707399368286,
      "epoch": 0.502236918887376,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497971979070776,
      "loss": 5.4289,
      "mean_token_accuracy": 0.1710171401500702,
      "num_tokens": 11204596.0,
      "step": 6455
    },
    {
      "entropy": 5.60586142539978,
      "epoch": 0.5026259482590936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979682645649834,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17194198518991471,
      "num_tokens": 11212629.0,
      "step": 6460
    },
    {
      "entropy": 5.6631279468536375,
      "epoch": 0.5030149776308112,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004979645466760025,
      "loss": 5.4461,
      "mean_token_accuracy": 0.1727239266037941,
      "num_tokens": 11221836.0,
      "step": 6465
    },
    {
      "entropy": 5.644039964675903,
      "epoch": 0.5034040070025287,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979608254038892,
      "loss": 5.5072,
      "mean_token_accuracy": 0.161077281832695,
      "num_tokens": 11230600.0,
      "step": 6470
    },
    {
      "entropy": 5.740381383895874,
      "epoch": 0.5037930363742462,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979571007487003,
      "loss": 5.4891,
      "mean_token_accuracy": 0.16057279258966445,
      "num_tokens": 11239664.0,
      "step": 6475
    },
    {
      "entropy": 5.589925527572632,
      "epoch": 0.5041820657459638,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979533727104924,
      "loss": 5.4073,
      "mean_token_accuracy": 0.1660124585032463,
      "num_tokens": 11249111.0,
      "step": 6480
    },
    {
      "entropy": 5.558521556854248,
      "epoch": 0.5045710951176814,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497949641289322,
      "loss": 5.3128,
      "mean_token_accuracy": 0.1722047045826912,
      "num_tokens": 11257599.0,
      "step": 6485
    },
    {
      "entropy": 5.5940399169921875,
      "epoch": 0.504960124489399,
      "grad_norm": 1.0,
      "learning_rate": 0.0004979459064852456,
      "loss": 5.433,
      "mean_token_accuracy": 0.18156720101833343,
      "num_tokens": 11265586.0,
      "step": 6490
    },
    {
      "entropy": 5.708671045303345,
      "epoch": 0.5053491538611166,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979421682983197,
      "loss": 5.5548,
      "mean_token_accuracy": 0.16661259680986404,
      "num_tokens": 11274736.0,
      "step": 6495
    },
    {
      "entropy": 5.630670738220215,
      "epoch": 0.505738183232834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979384267286015,
      "loss": 5.3417,
      "mean_token_accuracy": 0.18198635578155517,
      "num_tokens": 11282596.0,
      "step": 6500
    },
    {
      "entropy": 5.623176193237304,
      "epoch": 0.5061272126045516,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979346817761475,
      "loss": 5.4219,
      "mean_token_accuracy": 0.16544625163078308,
      "num_tokens": 11290974.0,
      "step": 6505
    },
    {
      "entropy": 5.506893968582153,
      "epoch": 0.5065162419762692,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979309334410144,
      "loss": 5.3031,
      "mean_token_accuracy": 0.17951635420322418,
      "num_tokens": 11299196.0,
      "step": 6510
    },
    {
      "entropy": 5.595285224914551,
      "epoch": 0.5069052713479868,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979271817232594,
      "loss": 5.4252,
      "mean_token_accuracy": 0.16717837750911713,
      "num_tokens": 11308214.0,
      "step": 6515
    },
    {
      "entropy": 5.613783502578736,
      "epoch": 0.5072943007197044,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979234266229391,
      "loss": 5.4134,
      "mean_token_accuracy": 0.17198467999696732,
      "num_tokens": 11316566.0,
      "step": 6520
    },
    {
      "entropy": 5.5896409034729,
      "epoch": 0.5076833300914219,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979196681401106,
      "loss": 5.3964,
      "mean_token_accuracy": 0.17299696058034897,
      "num_tokens": 11325079.0,
      "step": 6525
    },
    {
      "entropy": 5.591872882843018,
      "epoch": 0.5080723594631394,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497915906274831,
      "loss": 5.415,
      "mean_token_accuracy": 0.16943266242742538,
      "num_tokens": 11333829.0,
      "step": 6530
    },
    {
      "entropy": 5.550478506088257,
      "epoch": 0.508461388834857,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004979121410271574,
      "loss": 5.3552,
      "mean_token_accuracy": 0.1776174560189247,
      "num_tokens": 11343078.0,
      "step": 6535
    },
    {
      "entropy": 5.612886571884156,
      "epoch": 0.5088504182065746,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979083723971468,
      "loss": 5.4797,
      "mean_token_accuracy": 0.1714700222015381,
      "num_tokens": 11351475.0,
      "step": 6540
    },
    {
      "entropy": 5.579367780685425,
      "epoch": 0.5092394475782922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979046003848564,
      "loss": 5.3507,
      "mean_token_accuracy": 0.17437137365341188,
      "num_tokens": 11359810.0,
      "step": 6545
    },
    {
      "entropy": 5.637637901306152,
      "epoch": 0.5096284769500097,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979008249903435,
      "loss": 5.4018,
      "mean_token_accuracy": 0.17298720329999923,
      "num_tokens": 11368585.0,
      "step": 6550
    },
    {
      "entropy": 5.618145275115967,
      "epoch": 0.5100175063217273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978970462136655,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17616247981786728,
      "num_tokens": 11377393.0,
      "step": 6555
    },
    {
      "entropy": 5.609848976135254,
      "epoch": 0.5104065356934449,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978932640548794,
      "loss": 5.4073,
      "mean_token_accuracy": 0.165021151304245,
      "num_tokens": 11385877.0,
      "step": 6560
    },
    {
      "entropy": 5.553051280975342,
      "epoch": 0.5107955650651624,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004978894785140429,
      "loss": 5.3125,
      "mean_token_accuracy": 0.1768214926123619,
      "num_tokens": 11394457.0,
      "step": 6565
    },
    {
      "entropy": 5.530999422073364,
      "epoch": 0.51118459443688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978856895912132,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17647128850221633,
      "num_tokens": 11403007.0,
      "step": 6570
    },
    {
      "entropy": 5.6438335418701175,
      "epoch": 0.5115736238085975,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978818972864481,
      "loss": 5.4125,
      "mean_token_accuracy": 0.16737185269594193,
      "num_tokens": 11412237.0,
      "step": 6575
    },
    {
      "entropy": 5.6251890659332275,
      "epoch": 0.5119626531803151,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.3685,
      "mean_token_accuracy": 0.16872357279062272,
      "num_tokens": 11421219.0,
      "step": 6580
    },
    {
      "entropy": 5.618922996520996,
      "epoch": 0.5123516825520327,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978743025313413,
      "loss": 5.3626,
      "mean_token_accuracy": 0.16852298229932786,
      "num_tokens": 11428933.0,
      "step": 6585
    },
    {
      "entropy": 5.549186086654663,
      "epoch": 0.5127407119237503,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978705000811148,
      "loss": 5.4657,
      "mean_token_accuracy": 0.16887375563383103,
      "num_tokens": 11437551.0,
      "step": 6590
    },
    {
      "entropy": 5.6655247688293455,
      "epoch": 0.5131297412954678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978666942491832,
      "loss": 5.5484,
      "mean_token_accuracy": 0.16170087456703186,
      "num_tokens": 11446045.0,
      "step": 6595
    },
    {
      "entropy": 5.709177684783936,
      "epoch": 0.5135187706671853,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978628850356043,
      "loss": 5.4712,
      "mean_token_accuracy": 0.16923767775297166,
      "num_tokens": 11455159.0,
      "step": 6600
    },
    {
      "entropy": 5.635027027130127,
      "epoch": 0.5139078000389029,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978590724404358,
      "loss": 5.3689,
      "mean_token_accuracy": 0.17412376254796982,
      "num_tokens": 11463835.0,
      "step": 6605
    },
    {
      "entropy": 5.584535932540893,
      "epoch": 0.5142968294106205,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978552564637356,
      "loss": 5.4477,
      "mean_token_accuracy": 0.1742961212992668,
      "num_tokens": 11472439.0,
      "step": 6610
    },
    {
      "entropy": 5.613216876983643,
      "epoch": 0.5146858587823381,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004978514371055616,
      "loss": 5.482,
      "mean_token_accuracy": 0.1656497597694397,
      "num_tokens": 11481626.0,
      "step": 6615
    },
    {
      "entropy": 5.561394691467285,
      "epoch": 0.5150748881540557,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004978476143659718,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17834966033697128,
      "num_tokens": 11490571.0,
      "step": 6620
    },
    {
      "entropy": 5.6569291114807125,
      "epoch": 0.5154639175257731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978437882450241,
      "loss": 5.4625,
      "mean_token_accuracy": 0.1672450289130211,
      "num_tokens": 11499145.0,
      "step": 6625
    },
    {
      "entropy": 5.643627548217774,
      "epoch": 0.5158529468974907,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004978399587427766,
      "loss": 5.399,
      "mean_token_accuracy": 0.1676664710044861,
      "num_tokens": 11508133.0,
      "step": 6630
    },
    {
      "entropy": 5.459604072570801,
      "epoch": 0.5162419762692083,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978361258592874,
      "loss": 5.285,
      "mean_token_accuracy": 0.17830432951450348,
      "num_tokens": 11516790.0,
      "step": 6635
    },
    {
      "entropy": 5.564607620239258,
      "epoch": 0.5166310056409259,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004978322895946147,
      "loss": 5.3745,
      "mean_token_accuracy": 0.16994131356477737,
      "num_tokens": 11525963.0,
      "step": 6640
    },
    {
      "entropy": 5.590848112106324,
      "epoch": 0.5170200350126435,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978284499488167,
      "loss": 5.3554,
      "mean_token_accuracy": 0.17311257421970366,
      "num_tokens": 11534193.0,
      "step": 6645
    },
    {
      "entropy": 5.609114074707032,
      "epoch": 0.5174090643843611,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978246069219516,
      "loss": 5.375,
      "mean_token_accuracy": 0.16920317709445953,
      "num_tokens": 11542517.0,
      "step": 6650
    },
    {
      "entropy": 5.574249744415283,
      "epoch": 0.5177980937560785,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004978207605140777,
      "loss": 5.4459,
      "mean_token_accuracy": 0.16738450825214385,
      "num_tokens": 11550874.0,
      "step": 6655
    },
    {
      "entropy": 5.631093215942383,
      "epoch": 0.5181871231277961,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978169107252534,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16461260616779327,
      "num_tokens": 11560562.0,
      "step": 6660
    },
    {
      "entropy": 5.592811727523804,
      "epoch": 0.5185761524995137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978130575555373,
      "loss": 5.3429,
      "mean_token_accuracy": 0.17423924505710603,
      "num_tokens": 11569887.0,
      "step": 6665
    },
    {
      "entropy": 5.5215840339660645,
      "epoch": 0.5189651818712313,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978092010049877,
      "loss": 5.2705,
      "mean_token_accuracy": 0.18187254369258882,
      "num_tokens": 11578608.0,
      "step": 6670
    },
    {
      "entropy": 5.581983661651611,
      "epoch": 0.5193542112429489,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000497805341073663,
      "loss": 5.3912,
      "mean_token_accuracy": 0.17022958993911744,
      "num_tokens": 11587435.0,
      "step": 6675
    },
    {
      "entropy": 5.595023345947266,
      "epoch": 0.5197432406146664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978014777616219,
      "loss": 5.4862,
      "mean_token_accuracy": 0.16470641642808914,
      "num_tokens": 11596094.0,
      "step": 6680
    },
    {
      "entropy": 5.631581449508667,
      "epoch": 0.520132269986384,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004977976110689229,
      "loss": 5.4378,
      "mean_token_accuracy": 0.16952349245548248,
      "num_tokens": 11604521.0,
      "step": 6685
    },
    {
      "entropy": 5.641193103790283,
      "epoch": 0.5205212993581015,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000497793740995625,
      "loss": 5.4025,
      "mean_token_accuracy": 0.17098213583230973,
      "num_tokens": 11613143.0,
      "step": 6690
    },
    {
      "entropy": 5.630144119262695,
      "epoch": 0.5209103287298191,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977898675417866,
      "loss": 5.445,
      "mean_token_accuracy": 0.17159462794661523,
      "num_tokens": 11622667.0,
      "step": 6695
    },
    {
      "entropy": 5.602881956100464,
      "epoch": 0.5212993581015367,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977859907074664,
      "loss": 5.4167,
      "mean_token_accuracy": 0.1751892313361168,
      "num_tokens": 11632114.0,
      "step": 6700
    },
    {
      "entropy": 5.625202655792236,
      "epoch": 0.5216883874732542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977821104927236,
      "loss": 5.3335,
      "mean_token_accuracy": 0.17328906655311585,
      "num_tokens": 11640731.0,
      "step": 6705
    },
    {
      "entropy": 5.5024106979370115,
      "epoch": 0.5220774168449718,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004977782268976167,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1757868632674217,
      "num_tokens": 11648975.0,
      "step": 6710
    },
    {
      "entropy": 5.582853174209594,
      "epoch": 0.5224664462166894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977743399222048,
      "loss": 5.3494,
      "mean_token_accuracy": 0.17825236171483994,
      "num_tokens": 11657598.0,
      "step": 6715
    },
    {
      "entropy": 5.709061145782471,
      "epoch": 0.522855475588407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004977704495665471,
      "loss": 5.5124,
      "mean_token_accuracy": 0.1737692803144455,
      "num_tokens": 11665504.0,
      "step": 6720
    },
    {
      "entropy": 5.629387521743775,
      "epoch": 0.5232445049601245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977665558307023,
      "loss": 5.4329,
      "mean_token_accuracy": 0.17330906242132188,
      "num_tokens": 11674337.0,
      "step": 6725
    },
    {
      "entropy": 5.485547828674316,
      "epoch": 0.523633534331842,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977626587147295,
      "loss": 5.2716,
      "mean_token_accuracy": 0.18216868340969086,
      "num_tokens": 11682626.0,
      "step": 6730
    },
    {
      "entropy": 5.547606945037842,
      "epoch": 0.5240225637035596,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497758758218688,
      "loss": 5.3259,
      "mean_token_accuracy": 0.1781942442059517,
      "num_tokens": 11691403.0,
      "step": 6735
    },
    {
      "entropy": 5.651517009735107,
      "epoch": 0.5244115930752772,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977548543426368,
      "loss": 5.4553,
      "mean_token_accuracy": 0.166091488301754,
      "num_tokens": 11700590.0,
      "step": 6740
    },
    {
      "entropy": 5.625852966308594,
      "epoch": 0.5248006224469948,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977509470866353,
      "loss": 5.3641,
      "mean_token_accuracy": 0.16962600648403167,
      "num_tokens": 11708799.0,
      "step": 6745
    },
    {
      "entropy": 5.567790889739991,
      "epoch": 0.5251896518187124,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977470364507427,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17713444232940673,
      "num_tokens": 11716953.0,
      "step": 6750
    },
    {
      "entropy": 5.605804109573365,
      "epoch": 0.5255786811904298,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977431224350184,
      "loss": 5.4306,
      "mean_token_accuracy": 0.1685882940888405,
      "num_tokens": 11725032.0,
      "step": 6755
    },
    {
      "entropy": 5.613562965393067,
      "epoch": 0.5259677105621474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977392050395217,
      "loss": 5.4337,
      "mean_token_accuracy": 0.16917459070682525,
      "num_tokens": 11733359.0,
      "step": 6760
    },
    {
      "entropy": 5.665448999404907,
      "epoch": 0.526356739933865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977352842643121,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17600178420543672,
      "num_tokens": 11740936.0,
      "step": 6765
    },
    {
      "entropy": 5.61443829536438,
      "epoch": 0.5267457693055826,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497731360109449,
      "loss": 5.4559,
      "mean_token_accuracy": 0.16777198761701584,
      "num_tokens": 11749764.0,
      "step": 6770
    },
    {
      "entropy": 5.64350323677063,
      "epoch": 0.5271347986773002,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977274325749922,
      "loss": 5.4631,
      "mean_token_accuracy": 0.17370951920747757,
      "num_tokens": 11757945.0,
      "step": 6775
    },
    {
      "entropy": 5.547140598297119,
      "epoch": 0.5275238280490177,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977235016610011,
      "loss": 5.3372,
      "mean_token_accuracy": 0.17899322956800462,
      "num_tokens": 11766333.0,
      "step": 6780
    },
    {
      "entropy": 5.56026520729065,
      "epoch": 0.5279128574207352,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977195673675353,
      "loss": 5.3937,
      "mean_token_accuracy": 0.17261524647474288,
      "num_tokens": 11775041.0,
      "step": 6785
    },
    {
      "entropy": 5.610231876373291,
      "epoch": 0.5283018867924528,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977156296946545,
      "loss": 5.377,
      "mean_token_accuracy": 0.1705646276473999,
      "num_tokens": 11783958.0,
      "step": 6790
    },
    {
      "entropy": 5.579576539993286,
      "epoch": 0.5286909161641704,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977116886424187,
      "loss": 5.4475,
      "mean_token_accuracy": 0.16637502908706664,
      "num_tokens": 11792174.0,
      "step": 6795
    },
    {
      "entropy": 5.5774702548980715,
      "epoch": 0.529079945535888,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977077442108873,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17724884301424026,
      "num_tokens": 11800514.0,
      "step": 6800
    },
    {
      "entropy": 5.590536308288574,
      "epoch": 0.5294689749076055,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977037964001204,
      "loss": 5.2651,
      "mean_token_accuracy": 0.18220219165086746,
      "num_tokens": 11808918.0,
      "step": 6805
    },
    {
      "entropy": 5.654633092880249,
      "epoch": 0.5298580042793231,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497699845210178,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1643660172820091,
      "num_tokens": 11818014.0,
      "step": 6810
    },
    {
      "entropy": 5.5917236328125,
      "epoch": 0.5302470336510406,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004976958906411197,
      "loss": 5.4426,
      "mean_token_accuracy": 0.16338089257478713,
      "num_tokens": 11826477.0,
      "step": 6815
    },
    {
      "entropy": 5.673809576034546,
      "epoch": 0.5306360630227582,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004976919326930059,
      "loss": 5.4992,
      "mean_token_accuracy": 0.16627666354179382,
      "num_tokens": 11835513.0,
      "step": 6820
    },
    {
      "entropy": 5.695989656448364,
      "epoch": 0.5310250923944758,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976879713658964,
      "loss": 5.481,
      "mean_token_accuracy": 0.17325812727212905,
      "num_tokens": 11844495.0,
      "step": 6825
    },
    {
      "entropy": 5.628694486618042,
      "epoch": 0.5314141217661933,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004976840066598514,
      "loss": 5.4452,
      "mean_token_accuracy": 0.16433378607034682,
      "num_tokens": 11853013.0,
      "step": 6830
    },
    {
      "entropy": 5.51334285736084,
      "epoch": 0.5318031511379109,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976800385749309,
      "loss": 5.2287,
      "mean_token_accuracy": 0.17677582651376725,
      "num_tokens": 11861899.0,
      "step": 6835
    },
    {
      "entropy": 5.5065830707550045,
      "epoch": 0.5321921805096285,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976760671111954,
      "loss": 5.3119,
      "mean_token_accuracy": 0.18060407787561417,
      "num_tokens": 11870836.0,
      "step": 6840
    },
    {
      "entropy": 5.568042039871216,
      "epoch": 0.532581209881346,
      "grad_norm": 0.953125,
      "learning_rate": 0.000497672092268705,
      "loss": 5.3677,
      "mean_token_accuracy": 0.16628734916448593,
      "num_tokens": 11879363.0,
      "step": 6845
    },
    {
      "entropy": 5.6174729347229,
      "epoch": 0.5329702392530636,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049766811404752,
      "loss": 5.3655,
      "mean_token_accuracy": 0.17461650520563127,
      "num_tokens": 11887696.0,
      "step": 6850
    },
    {
      "entropy": 5.585926151275634,
      "epoch": 0.5333592686247812,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976641324477008,
      "loss": 5.3776,
      "mean_token_accuracy": 0.1717255249619484,
      "num_tokens": 11895867.0,
      "step": 6855
    },
    {
      "entropy": 5.632009267807007,
      "epoch": 0.5337482979964987,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976601474693077,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17238852828741075,
      "num_tokens": 11904537.0,
      "step": 6860
    },
    {
      "entropy": 5.675828123092652,
      "epoch": 0.5341373273682163,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976561591124014,
      "loss": 5.4333,
      "mean_token_accuracy": 0.17629104405641555,
      "num_tokens": 11913097.0,
      "step": 6865
    },
    {
      "entropy": 5.496969223022461,
      "epoch": 0.5345263567399339,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004976521673770421,
      "loss": 5.2426,
      "mean_token_accuracy": 0.17388982623815535,
      "num_tokens": 11921852.0,
      "step": 6870
    },
    {
      "entropy": 5.594495582580566,
      "epoch": 0.5349153861116515,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004976481722632907,
      "loss": 5.427,
      "mean_token_accuracy": 0.1668204829096794,
      "num_tokens": 11930836.0,
      "step": 6875
    },
    {
      "entropy": 5.647954893112183,
      "epoch": 0.535304415483369,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976441737712076,
      "loss": 5.4943,
      "mean_token_accuracy": 0.1578487128019333,
      "num_tokens": 11940069.0,
      "step": 6880
    },
    {
      "entropy": 5.67997374534607,
      "epoch": 0.5356934448550865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976401719008536,
      "loss": 5.4275,
      "mean_token_accuracy": 0.17309840321540831,
      "num_tokens": 11948525.0,
      "step": 6885
    },
    {
      "entropy": 5.600617599487305,
      "epoch": 0.5360824742268041,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976361666522893,
      "loss": 5.4058,
      "mean_token_accuracy": 0.17392354756593703,
      "num_tokens": 11956912.0,
      "step": 6890
    },
    {
      "entropy": 5.507595109939575,
      "epoch": 0.5364715035985217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976321580255755,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16616453230381012,
      "num_tokens": 11965782.0,
      "step": 6895
    },
    {
      "entropy": 5.623979187011718,
      "epoch": 0.5368605329702393,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004976281460207731,
      "loss": 5.4132,
      "mean_token_accuracy": 0.17315759211778642,
      "num_tokens": 11974889.0,
      "step": 6900
    },
    {
      "entropy": 5.655584335327148,
      "epoch": 0.5372495623419569,
      "grad_norm": 0.9375,
      "learning_rate": 0.000497624130637943,
      "loss": 5.4103,
      "mean_token_accuracy": 0.1678052067756653,
      "num_tokens": 11983507.0,
      "step": 6905
    },
    {
      "entropy": 5.534413576126099,
      "epoch": 0.5376385917136743,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497620111877146,
      "loss": 5.275,
      "mean_token_accuracy": 0.1823041096329689,
      "num_tokens": 11991646.0,
      "step": 6910
    },
    {
      "entropy": 5.4777453422546385,
      "epoch": 0.5380276210853919,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976160897384431,
      "loss": 5.3216,
      "mean_token_accuracy": 0.17367151379585266,
      "num_tokens": 12000223.0,
      "step": 6915
    },
    {
      "entropy": 5.622479200363159,
      "epoch": 0.5384166504571095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976120642218955,
      "loss": 5.4153,
      "mean_token_accuracy": 0.17153375297784806,
      "num_tokens": 12009012.0,
      "step": 6920
    },
    {
      "entropy": 5.717148637771606,
      "epoch": 0.5388056798288271,
      "grad_norm": 0.953125,
      "learning_rate": 0.000497608035327564,
      "loss": 5.459,
      "mean_token_accuracy": 0.16472576260566713,
      "num_tokens": 12018001.0,
      "step": 6925
    },
    {
      "entropy": 5.5548868656158445,
      "epoch": 0.5391947092005447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049760400305551,
      "loss": 5.2903,
      "mean_token_accuracy": 0.18364150524139405,
      "num_tokens": 12025943.0,
      "step": 6930
    },
    {
      "entropy": 5.576215124130249,
      "epoch": 0.5395837385722622,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004975999674057944,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1743236228823662,
      "num_tokens": 12035073.0,
      "step": 6935
    },
    {
      "entropy": 5.650489425659179,
      "epoch": 0.5399727679439797,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004975959283784787,
      "loss": 5.5105,
      "mean_token_accuracy": 0.16534960120916367,
      "num_tokens": 12043698.0,
      "step": 6940
    },
    {
      "entropy": 5.6009152889251705,
      "epoch": 0.5403617973156973,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975918859736241,
      "loss": 5.3966,
      "mean_token_accuracy": 0.17727003395557403,
      "num_tokens": 12052247.0,
      "step": 6945
    },
    {
      "entropy": 5.594502592086792,
      "epoch": 0.5407508266874149,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975878401912919,
      "loss": 5.4101,
      "mean_token_accuracy": 0.1762482464313507,
      "num_tokens": 12061187.0,
      "step": 6950
    },
    {
      "entropy": 5.57668399810791,
      "epoch": 0.5411398560591325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975837910315436,
      "loss": 5.3585,
      "mean_token_accuracy": 0.170804201066494,
      "num_tokens": 12070121.0,
      "step": 6955
    },
    {
      "entropy": 5.609505748748779,
      "epoch": 0.54152888543085,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975797384944405,
      "loss": 5.3086,
      "mean_token_accuracy": 0.18491584211587905,
      "num_tokens": 12078578.0,
      "step": 6960
    },
    {
      "entropy": 5.585539960861206,
      "epoch": 0.5419179148025676,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497575682580044,
      "loss": 5.466,
      "mean_token_accuracy": 0.1712225630879402,
      "num_tokens": 12087029.0,
      "step": 6965
    },
    {
      "entropy": 5.7061645030975345,
      "epoch": 0.5423069441742852,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000497571623288416,
      "loss": 5.435,
      "mean_token_accuracy": 0.16743697375059127,
      "num_tokens": 12096105.0,
      "step": 6970
    },
    {
      "entropy": 5.756121730804443,
      "epoch": 0.5426959735460027,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975675606196177,
      "loss": 5.506,
      "mean_token_accuracy": 0.16332751214504243,
      "num_tokens": 12105098.0,
      "step": 6975
    },
    {
      "entropy": 5.599440097808838,
      "epoch": 0.5430850029177203,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497563494573711,
      "loss": 5.3063,
      "mean_token_accuracy": 0.17674217969179154,
      "num_tokens": 12112894.0,
      "step": 6980
    },
    {
      "entropy": 5.5193421840667725,
      "epoch": 0.5434740322894378,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975594251507575,
      "loss": 5.4403,
      "mean_token_accuracy": 0.17357636839151383,
      "num_tokens": 12121457.0,
      "step": 6985
    },
    {
      "entropy": 5.639322137832641,
      "epoch": 0.5438630616611554,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497555352350819,
      "loss": 5.3172,
      "mean_token_accuracy": 0.1768542766571045,
      "num_tokens": 12129998.0,
      "step": 6990
    },
    {
      "entropy": 5.666340065002442,
      "epoch": 0.544252091032873,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975512761739572,
      "loss": 5.389,
      "mean_token_accuracy": 0.17072022557258607,
      "num_tokens": 12137861.0,
      "step": 6995
    },
    {
      "entropy": 5.556141805648804,
      "epoch": 0.5446411204045906,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497547196620234,
      "loss": 5.3912,
      "mean_token_accuracy": 0.17440045475959778,
      "num_tokens": 12146294.0,
      "step": 7000
    },
    {
      "entropy": 5.559613847732544,
      "epoch": 0.5450301497763081,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975431136897114,
      "loss": 5.3929,
      "mean_token_accuracy": 0.17047546803951263,
      "num_tokens": 12155642.0,
      "step": 7005
    },
    {
      "entropy": 5.735659790039063,
      "epoch": 0.5454191791480256,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975390273824512,
      "loss": 5.4939,
      "mean_token_accuracy": 0.16992603689432145,
      "num_tokens": 12164709.0,
      "step": 7010
    },
    {
      "entropy": 5.617656564712524,
      "epoch": 0.5458082085197432,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975349376985155,
      "loss": 5.3662,
      "mean_token_accuracy": 0.17637594640254975,
      "num_tokens": 12173591.0,
      "step": 7015
    },
    {
      "entropy": 5.585110187530518,
      "epoch": 0.5461972378914608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975308446379663,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1666642650961876,
      "num_tokens": 12182050.0,
      "step": 7020
    },
    {
      "entropy": 5.649959850311279,
      "epoch": 0.5465862672631784,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004975267482008657,
      "loss": 5.4722,
      "mean_token_accuracy": 0.16861061453819276,
      "num_tokens": 12190870.0,
      "step": 7025
    },
    {
      "entropy": 5.492293834686279,
      "epoch": 0.546975296634896,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497522648387276,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18491959273815156,
      "num_tokens": 12199624.0,
      "step": 7030
    },
    {
      "entropy": 5.6087899684906,
      "epoch": 0.5473643260066134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975185451972592,
      "loss": 5.4624,
      "mean_token_accuracy": 0.16690078377723694,
      "num_tokens": 12208087.0,
      "step": 7035
    },
    {
      "entropy": 5.680318450927734,
      "epoch": 0.547753355378331,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975144386308776,
      "loss": 5.4131,
      "mean_token_accuracy": 0.17517598122358322,
      "num_tokens": 12216531.0,
      "step": 7040
    },
    {
      "entropy": 5.615996551513672,
      "epoch": 0.5481423847500486,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975103286881936,
      "loss": 5.3705,
      "mean_token_accuracy": 0.17713344991207122,
      "num_tokens": 12225094.0,
      "step": 7045
    },
    {
      "entropy": 5.560548114776611,
      "epoch": 0.5485314141217662,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004975062153692696,
      "loss": 5.3765,
      "mean_token_accuracy": 0.17883004546165465,
      "num_tokens": 12233824.0,
      "step": 7050
    },
    {
      "entropy": 5.554152297973633,
      "epoch": 0.5489204434934838,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004975020986741678,
      "loss": 5.2767,
      "mean_token_accuracy": 0.17529497891664506,
      "num_tokens": 12242923.0,
      "step": 7055
    },
    {
      "entropy": 5.581981992721557,
      "epoch": 0.5493094728652014,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974979786029509,
      "loss": 5.3156,
      "mean_token_accuracy": 0.17031585723161696,
      "num_tokens": 12251963.0,
      "step": 7060
    },
    {
      "entropy": 5.586628723144531,
      "epoch": 0.5496985022369189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974938551556814,
      "loss": 5.4442,
      "mean_token_accuracy": 0.17018938660621644,
      "num_tokens": 12260347.0,
      "step": 7065
    },
    {
      "entropy": 5.551030397415161,
      "epoch": 0.5500875316086364,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974897283324217,
      "loss": 5.4813,
      "mean_token_accuracy": 0.16576917320489884,
      "num_tokens": 12269156.0,
      "step": 7070
    },
    {
      "entropy": 5.632766628265381,
      "epoch": 0.550476560980354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974855981332343,
      "loss": 5.3835,
      "mean_token_accuracy": 0.17017189860343934,
      "num_tokens": 12277636.0,
      "step": 7075
    },
    {
      "entropy": 5.631530666351319,
      "epoch": 0.5508655903520716,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974814645581823,
      "loss": 5.3576,
      "mean_token_accuracy": 0.17621888518333434,
      "num_tokens": 12286027.0,
      "step": 7080
    },
    {
      "entropy": 5.633374929428101,
      "epoch": 0.5512546197237892,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974773276073282,
      "loss": 5.4807,
      "mean_token_accuracy": 0.16844650804996492,
      "num_tokens": 12295085.0,
      "step": 7085
    },
    {
      "entropy": 5.676600170135498,
      "epoch": 0.5516436490955067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974731872807347,
      "loss": 5.4533,
      "mean_token_accuracy": 0.17263481616973878,
      "num_tokens": 12304275.0,
      "step": 7090
    },
    {
      "entropy": 5.550247097015381,
      "epoch": 0.5520326784672243,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004974690435784647,
      "loss": 5.2629,
      "mean_token_accuracy": 0.1946805715560913,
      "num_tokens": 12312841.0,
      "step": 7095
    },
    {
      "entropy": 5.572363948822021,
      "epoch": 0.5524217078389418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974648965005811,
      "loss": 5.3797,
      "mean_token_accuracy": 0.16614917069673538,
      "num_tokens": 12322133.0,
      "step": 7100
    },
    {
      "entropy": 5.581380891799927,
      "epoch": 0.5528107372106594,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974607460471466,
      "loss": 5.3535,
      "mean_token_accuracy": 0.17336255609989165,
      "num_tokens": 12330502.0,
      "step": 7105
    },
    {
      "entropy": 5.6251250267028805,
      "epoch": 0.553199766582377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974565922182246,
      "loss": 5.3795,
      "mean_token_accuracy": 0.17468350678682326,
      "num_tokens": 12338899.0,
      "step": 7110
    },
    {
      "entropy": 5.556831884384155,
      "epoch": 0.5535887959540945,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974524350138777,
      "loss": 5.3728,
      "mean_token_accuracy": 0.17247378528118135,
      "num_tokens": 12348299.0,
      "step": 7115
    },
    {
      "entropy": 5.624515962600708,
      "epoch": 0.5539778253258121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974482744341693,
      "loss": 5.3927,
      "mean_token_accuracy": 0.17165566980838776,
      "num_tokens": 12356582.0,
      "step": 7120
    },
    {
      "entropy": 5.673182439804077,
      "epoch": 0.5543668546975297,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974441104791624,
      "loss": 5.4972,
      "mean_token_accuracy": 0.15884086936712266,
      "num_tokens": 12366209.0,
      "step": 7125
    },
    {
      "entropy": 5.608900737762451,
      "epoch": 0.5547558840692473,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974399431489201,
      "loss": 5.4221,
      "mean_token_accuracy": 0.16073356568813324,
      "num_tokens": 12374777.0,
      "step": 7130
    },
    {
      "entropy": 5.546727514266967,
      "epoch": 0.5551449134409648,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974357724435057,
      "loss": 5.381,
      "mean_token_accuracy": 0.16626832336187364,
      "num_tokens": 12384405.0,
      "step": 7135
    },
    {
      "entropy": 5.6791276931762695,
      "epoch": 0.5555339428126823,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004974315983629825,
      "loss": 5.4615,
      "mean_token_accuracy": 0.16987256854772567,
      "num_tokens": 12393464.0,
      "step": 7140
    },
    {
      "entropy": 5.695638513565063,
      "epoch": 0.5559229721843999,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004974274209074139,
      "loss": 5.4619,
      "mean_token_accuracy": 0.1698928251862526,
      "num_tokens": 12401917.0,
      "step": 7145
    },
    {
      "entropy": 5.474405193328858,
      "epoch": 0.5563120015561175,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974232400768632,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1800903260707855,
      "num_tokens": 12410928.0,
      "step": 7150
    },
    {
      "entropy": 5.548608446121216,
      "epoch": 0.5567010309278351,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004974190558713939,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17504773437976837,
      "num_tokens": 12419747.0,
      "step": 7155
    },
    {
      "entropy": 5.63728699684143,
      "epoch": 0.5570900602995527,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004974148682910694,
      "loss": 5.364,
      "mean_token_accuracy": 0.17363888621330262,
      "num_tokens": 12428608.0,
      "step": 7160
    },
    {
      "entropy": 5.69043550491333,
      "epoch": 0.5574790896712701,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974106773359533,
      "loss": 5.4498,
      "mean_token_accuracy": 0.16662293970584868,
      "num_tokens": 12437101.0,
      "step": 7165
    },
    {
      "entropy": 5.541791868209839,
      "epoch": 0.5578681190429877,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974064830061091,
      "loss": 5.4083,
      "mean_token_accuracy": 0.16861678659915924,
      "num_tokens": 12446813.0,
      "step": 7170
    },
    {
      "entropy": 5.622215843200683,
      "epoch": 0.5582571484147053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974022853016006,
      "loss": 5.3703,
      "mean_token_accuracy": 0.17372025400400162,
      "num_tokens": 12455931.0,
      "step": 7175
    },
    {
      "entropy": 5.6301921844482425,
      "epoch": 0.5586461777864229,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973980842224913,
      "loss": 5.3379,
      "mean_token_accuracy": 0.17587753683328627,
      "num_tokens": 12463586.0,
      "step": 7180
    },
    {
      "entropy": 5.559577131271363,
      "epoch": 0.5590352071581405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973938797688452,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17821425795555115,
      "num_tokens": 12472063.0,
      "step": 7185
    },
    {
      "entropy": 5.597116994857788,
      "epoch": 0.559424236529858,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973896719407259,
      "loss": 5.3745,
      "mean_token_accuracy": 0.17003683000802994,
      "num_tokens": 12480763.0,
      "step": 7190
    },
    {
      "entropy": 5.614082002639771,
      "epoch": 0.5598132659015755,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973854607381972,
      "loss": 5.3305,
      "mean_token_accuracy": 0.1768320009112358,
      "num_tokens": 12488757.0,
      "step": 7195
    },
    {
      "entropy": 5.5355589389801025,
      "epoch": 0.5602022952732931,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973812461613232,
      "loss": 5.3783,
      "mean_token_accuracy": 0.17371366620063783,
      "num_tokens": 12497030.0,
      "step": 7200
    },
    {
      "entropy": 5.5868391513824465,
      "epoch": 0.5605913246450107,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973770282101677,
      "loss": 5.397,
      "mean_token_accuracy": 0.1695732444524765,
      "num_tokens": 12506067.0,
      "step": 7205
    },
    {
      "entropy": 5.5814941883087155,
      "epoch": 0.5609803540167283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973728068847947,
      "loss": 5.3318,
      "mean_token_accuracy": 0.1736765220761299,
      "num_tokens": 12514246.0,
      "step": 7210
    },
    {
      "entropy": 5.609518241882324,
      "epoch": 0.5613693833884458,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973685821852684,
      "loss": 5.3719,
      "mean_token_accuracy": 0.17128328680992128,
      "num_tokens": 12523523.0,
      "step": 7215
    },
    {
      "entropy": 5.615756368637085,
      "epoch": 0.5617584127601634,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973643541116528,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17098272442817689,
      "num_tokens": 12532142.0,
      "step": 7220
    },
    {
      "entropy": 5.577594757080078,
      "epoch": 0.562147442131881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497360122664012,
      "loss": 5.4065,
      "mean_token_accuracy": 0.16962229311466218,
      "num_tokens": 12540647.0,
      "step": 7225
    },
    {
      "entropy": 5.551422119140625,
      "epoch": 0.5625364715035985,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973558878424104,
      "loss": 5.3217,
      "mean_token_accuracy": 0.17154066264629364,
      "num_tokens": 12549010.0,
      "step": 7230
    },
    {
      "entropy": 5.5718786239624025,
      "epoch": 0.5629255008753161,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004973516496469119,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17378498762845992,
      "num_tokens": 12557428.0,
      "step": 7235
    },
    {
      "entropy": 5.51436676979065,
      "epoch": 0.5633145302470336,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004973474080775811,
      "loss": 5.2595,
      "mean_token_accuracy": 0.17589602768421173,
      "num_tokens": 12566610.0,
      "step": 7240
    },
    {
      "entropy": 5.515880918502807,
      "epoch": 0.5637035596187512,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973431631344824,
      "loss": 5.3742,
      "mean_token_accuracy": 0.17670979499816894,
      "num_tokens": 12576016.0,
      "step": 7245
    },
    {
      "entropy": 5.48199200630188,
      "epoch": 0.5640925889904688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049733891481768,
      "loss": 5.2015,
      "mean_token_accuracy": 0.18723199367523194,
      "num_tokens": 12584033.0,
      "step": 7250
    },
    {
      "entropy": 5.607383584976196,
      "epoch": 0.5644816183621864,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973346631272384,
      "loss": 5.3339,
      "mean_token_accuracy": 0.1781315878033638,
      "num_tokens": 12592784.0,
      "step": 7255
    },
    {
      "entropy": 5.615625619888306,
      "epoch": 0.5648706477339039,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973304080632222,
      "loss": 5.4719,
      "mean_token_accuracy": 0.1671131283044815,
      "num_tokens": 12601259.0,
      "step": 7260
    },
    {
      "entropy": 5.5404987812042235,
      "epoch": 0.5652596771056215,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497326149625696,
      "loss": 5.3057,
      "mean_token_accuracy": 0.16903193295001984,
      "num_tokens": 12610264.0,
      "step": 7265
    },
    {
      "entropy": 5.611272859573364,
      "epoch": 0.565648706477339,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004973218878147244,
      "loss": 5.312,
      "mean_token_accuracy": 0.17460647821426392,
      "num_tokens": 12618817.0,
      "step": 7270
    },
    {
      "entropy": 5.547258949279785,
      "epoch": 0.5660377358490566,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004973176226303719,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17069429606199266,
      "num_tokens": 12627224.0,
      "step": 7275
    },
    {
      "entropy": 5.619957113265992,
      "epoch": 0.5664267652207742,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973133540727033,
      "loss": 5.3435,
      "mean_token_accuracy": 0.1835724949836731,
      "num_tokens": 12635020.0,
      "step": 7280
    },
    {
      "entropy": 5.6189040184021,
      "epoch": 0.5668157945924918,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973090821417835,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1689133107662201,
      "num_tokens": 12643821.0,
      "step": 7285
    },
    {
      "entropy": 5.663434267044067,
      "epoch": 0.5672048239642093,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973048068376772,
      "loss": 5.4172,
      "mean_token_accuracy": 0.16706288903951644,
      "num_tokens": 12652091.0,
      "step": 7290
    },
    {
      "entropy": 5.639904451370239,
      "epoch": 0.5675938533359268,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973005281604492,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17488816380500793,
      "num_tokens": 12661389.0,
      "step": 7295
    },
    {
      "entropy": 5.608521747589111,
      "epoch": 0.5679828827076444,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972962461101646,
      "loss": 5.4461,
      "mean_token_accuracy": 0.17284394204616546,
      "num_tokens": 12670361.0,
      "step": 7300
    },
    {
      "entropy": 5.547716951370239,
      "epoch": 0.568371912079362,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972919606868883,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1760653153061867,
      "num_tokens": 12677765.0,
      "step": 7305
    },
    {
      "entropy": 5.5433228492736815,
      "epoch": 0.5687609414510796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972876718906852,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18404784053564072,
      "num_tokens": 12686461.0,
      "step": 7310
    },
    {
      "entropy": 5.5584510326385494,
      "epoch": 0.5691499708227972,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972833797216206,
      "loss": 5.3116,
      "mean_token_accuracy": 0.17475184798240662,
      "num_tokens": 12694663.0,
      "step": 7315
    },
    {
      "entropy": 5.576896142959595,
      "epoch": 0.5695390001945146,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004972790841797595,
      "loss": 5.3838,
      "mean_token_accuracy": 0.17167252004146577,
      "num_tokens": 12703659.0,
      "step": 7320
    },
    {
      "entropy": 5.546174049377441,
      "epoch": 0.5699280295662322,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497274785265167,
      "loss": 5.3451,
      "mean_token_accuracy": 0.17827020734548568,
      "num_tokens": 12711402.0,
      "step": 7325
    },
    {
      "entropy": 5.593303918838501,
      "epoch": 0.5703170589379498,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004972704829779086,
      "loss": 5.4083,
      "mean_token_accuracy": 0.16783702671527861,
      "num_tokens": 12720500.0,
      "step": 7330
    },
    {
      "entropy": 5.664913368225098,
      "epoch": 0.5707060883096674,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972661773180492,
      "loss": 5.4369,
      "mean_token_accuracy": 0.1647883877158165,
      "num_tokens": 12729325.0,
      "step": 7335
    },
    {
      "entropy": 5.6450872898101805,
      "epoch": 0.571095117681385,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972618682856545,
      "loss": 5.4592,
      "mean_token_accuracy": 0.16735735088586806,
      "num_tokens": 12738573.0,
      "step": 7340
    },
    {
      "entropy": 5.566001844406128,
      "epoch": 0.5714841470531025,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972575558807896,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17694940716028212,
      "num_tokens": 12746608.0,
      "step": 7345
    },
    {
      "entropy": 5.562113475799561,
      "epoch": 0.57187317642482,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972532401035201,
      "loss": 5.3049,
      "mean_token_accuracy": 0.17716436237096786,
      "num_tokens": 12754724.0,
      "step": 7350
    },
    {
      "entropy": 5.545844411849975,
      "epoch": 0.5722622057965376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972489209539114,
      "loss": 5.3816,
      "mean_token_accuracy": 0.1792597457766533,
      "num_tokens": 12762654.0,
      "step": 7355
    },
    {
      "entropy": 5.579734849929809,
      "epoch": 0.5726512351682552,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004972445984320292,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16492784172296523,
      "num_tokens": 12771615.0,
      "step": 7360
    },
    {
      "entropy": 5.691576433181763,
      "epoch": 0.5730402645399728,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972402725379388,
      "loss": 5.4626,
      "mean_token_accuracy": 0.17010326236486434,
      "num_tokens": 12780991.0,
      "step": 7365
    },
    {
      "entropy": 5.644831895828247,
      "epoch": 0.5734292939116903,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497235943271706,
      "loss": 5.3302,
      "mean_token_accuracy": 0.17996573746204375,
      "num_tokens": 12789402.0,
      "step": 7370
    },
    {
      "entropy": 5.5373939037322994,
      "epoch": 0.5738183232834079,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972316106333966,
      "loss": 5.3083,
      "mean_token_accuracy": 0.1795029029250145,
      "num_tokens": 12798095.0,
      "step": 7375
    },
    {
      "entropy": 5.63902177810669,
      "epoch": 0.5742073526551255,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972272746230761,
      "loss": 5.4044,
      "mean_token_accuracy": 0.17136958986520767,
      "num_tokens": 12806976.0,
      "step": 7380
    },
    {
      "entropy": 5.585088205337525,
      "epoch": 0.574596382026843,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004972229352408104,
      "loss": 5.3162,
      "mean_token_accuracy": 0.18278878778219224,
      "num_tokens": 12815228.0,
      "step": 7385
    },
    {
      "entropy": 5.585175561904907,
      "epoch": 0.5749854113985606,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972185924866655,
      "loss": 5.4097,
      "mean_token_accuracy": 0.16959019750356674,
      "num_tokens": 12823674.0,
      "step": 7390
    },
    {
      "entropy": 5.6082837104797365,
      "epoch": 0.5753744407702781,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972142463607071,
      "loss": 5.512,
      "mean_token_accuracy": 0.16803084313869476,
      "num_tokens": 12832810.0,
      "step": 7395
    },
    {
      "entropy": 5.687703227996826,
      "epoch": 0.5757634701419957,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004972098968630012,
      "loss": 5.4418,
      "mean_token_accuracy": 0.16590020209550857,
      "num_tokens": 12841430.0,
      "step": 7400
    },
    {
      "entropy": 5.609046506881714,
      "epoch": 0.5761524995137133,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972055439936137,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17771490216255187,
      "num_tokens": 12850539.0,
      "step": 7405
    },
    {
      "entropy": 5.6279497146606445,
      "epoch": 0.5765415288854309,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972011877526109,
      "loss": 5.447,
      "mean_token_accuracy": 0.17289159595966339,
      "num_tokens": 12859066.0,
      "step": 7410
    },
    {
      "entropy": 5.52370662689209,
      "epoch": 0.5769305582571485,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971968281400587,
      "loss": 5.3349,
      "mean_token_accuracy": 0.18174415230751037,
      "num_tokens": 12867603.0,
      "step": 7415
    },
    {
      "entropy": 5.455781126022339,
      "epoch": 0.5773195876288659,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971924651560233,
      "loss": 5.2013,
      "mean_token_accuracy": 0.18343908637762069,
      "num_tokens": 12875950.0,
      "step": 7420
    },
    {
      "entropy": 5.640551328659058,
      "epoch": 0.5777086170005835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004971880988005709,
      "loss": 5.4606,
      "mean_token_accuracy": 0.16838063895702363,
      "num_tokens": 12884154.0,
      "step": 7425
    },
    {
      "entropy": 5.654052639007569,
      "epoch": 0.5780976463723011,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971837290737677,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16462810784578324,
      "num_tokens": 12892515.0,
      "step": 7430
    },
    {
      "entropy": 5.546605968475342,
      "epoch": 0.5784866757440187,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971793559756801,
      "loss": 5.322,
      "mean_token_accuracy": 0.18057667464017868,
      "num_tokens": 12900919.0,
      "step": 7435
    },
    {
      "entropy": 5.5282165050506595,
      "epoch": 0.5788757051157363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004971749795063744,
      "loss": 5.2933,
      "mean_token_accuracy": 0.18205010741949082,
      "num_tokens": 12909338.0,
      "step": 7440
    },
    {
      "entropy": 5.652127838134765,
      "epoch": 0.5792647344874537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497170599665917,
      "loss": 5.4062,
      "mean_token_accuracy": 0.1692366525530815,
      "num_tokens": 12918118.0,
      "step": 7445
    },
    {
      "entropy": 5.635789680480957,
      "epoch": 0.5796537638591713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971662164543743,
      "loss": 5.4427,
      "mean_token_accuracy": 0.16868827491998672,
      "num_tokens": 12926878.0,
      "step": 7450
    },
    {
      "entropy": 5.559504508972168,
      "epoch": 0.5800427932308889,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971618298718131,
      "loss": 5.3067,
      "mean_token_accuracy": 0.17664028257131575,
      "num_tokens": 12934869.0,
      "step": 7455
    },
    {
      "entropy": 5.508236598968506,
      "epoch": 0.5804318226026065,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971574399182995,
      "loss": 5.3606,
      "mean_token_accuracy": 0.17187800258398056,
      "num_tokens": 12944281.0,
      "step": 7460
    },
    {
      "entropy": 5.611026191711426,
      "epoch": 0.5808208519743241,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004971530465939005,
      "loss": 5.3873,
      "mean_token_accuracy": 0.17068445384502412,
      "num_tokens": 12953281.0,
      "step": 7465
    },
    {
      "entropy": 5.613557481765747,
      "epoch": 0.5812098813460417,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971486498986825,
      "loss": 5.3671,
      "mean_token_accuracy": 0.1724362462759018,
      "num_tokens": 12962136.0,
      "step": 7470
    },
    {
      "entropy": 5.509874391555786,
      "epoch": 0.5815989107177592,
      "grad_norm": 1.0,
      "learning_rate": 0.0004971442498327124,
      "loss": 5.3246,
      "mean_token_accuracy": 0.17531097084283828,
      "num_tokens": 12971288.0,
      "step": 7475
    },
    {
      "entropy": 5.574598217010498,
      "epoch": 0.5819879400894767,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971398463960568,
      "loss": 5.2967,
      "mean_token_accuracy": 0.1769063279032707,
      "num_tokens": 12979836.0,
      "step": 7480
    },
    {
      "entropy": 5.564197444915772,
      "epoch": 0.5823769694611943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971354395887827,
      "loss": 5.3059,
      "mean_token_accuracy": 0.18093591332435607,
      "num_tokens": 12988233.0,
      "step": 7485
    },
    {
      "entropy": 5.561965847015381,
      "epoch": 0.5827659988329119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971310294109569,
      "loss": 5.317,
      "mean_token_accuracy": 0.17434777021408082,
      "num_tokens": 12997077.0,
      "step": 7490
    },
    {
      "entropy": 5.549249458312988,
      "epoch": 0.5831550282046295,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004971266158626461,
      "loss": 5.3422,
      "mean_token_accuracy": 0.17838245630264282,
      "num_tokens": 13005058.0,
      "step": 7495
    },
    {
      "entropy": 5.6491687297821045,
      "epoch": 0.583544057576347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004971221989439177,
      "loss": 5.4007,
      "mean_token_accuracy": 0.17415667623281478,
      "num_tokens": 13013347.0,
      "step": 7500
    },
    {
      "entropy": 5.656944322586059,
      "epoch": 0.5839330869480646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004971177786548384,
      "loss": 5.3388,
      "mean_token_accuracy": 0.1761828675866127,
      "num_tokens": 13021555.0,
      "step": 7505
    },
    {
      "entropy": 5.568319988250733,
      "epoch": 0.5843221163197821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971133549954753,
      "loss": 5.414,
      "mean_token_accuracy": 0.17406437993049623,
      "num_tokens": 13029817.0,
      "step": 7510
    },
    {
      "entropy": 5.583380794525146,
      "epoch": 0.5847111456914997,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971089279658954,
      "loss": 5.3826,
      "mean_token_accuracy": 0.17224892526865004,
      "num_tokens": 13038119.0,
      "step": 7515
    },
    {
      "entropy": 5.566053342819214,
      "epoch": 0.5851001750632173,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971044975661663,
      "loss": 5.31,
      "mean_token_accuracy": 0.17912320643663407,
      "num_tokens": 13047004.0,
      "step": 7520
    },
    {
      "entropy": 5.577041339874268,
      "epoch": 0.5854892044349348,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004971000637963549,
      "loss": 5.2967,
      "mean_token_accuracy": 0.1811252474784851,
      "num_tokens": 13055045.0,
      "step": 7525
    },
    {
      "entropy": 5.614419794082641,
      "epoch": 0.5858782338066524,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004970956266565285,
      "loss": 5.3933,
      "mean_token_accuracy": 0.17078788280487062,
      "num_tokens": 13064535.0,
      "step": 7530
    },
    {
      "entropy": 5.615678119659424,
      "epoch": 0.58626726317837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970911861467544,
      "loss": 5.4565,
      "mean_token_accuracy": 0.16273304224014282,
      "num_tokens": 13073089.0,
      "step": 7535
    },
    {
      "entropy": 5.569165563583374,
      "epoch": 0.5866562925500876,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004970867422671002,
      "loss": 5.3926,
      "mean_token_accuracy": 0.17114685624837875,
      "num_tokens": 13082161.0,
      "step": 7540
    },
    {
      "entropy": 5.559341335296631,
      "epoch": 0.5870453219218051,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000497082295017633,
      "loss": 5.261,
      "mean_token_accuracy": 0.18079383820295333,
      "num_tokens": 13090632.0,
      "step": 7545
    },
    {
      "entropy": 5.624056482315064,
      "epoch": 0.5874343512935226,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970778443984206,
      "loss": 5.3232,
      "mean_token_accuracy": 0.17447323203086854,
      "num_tokens": 13098825.0,
      "step": 7550
    },
    {
      "entropy": 5.634274244308472,
      "epoch": 0.5878233806652402,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004970733904095303,
      "loss": 5.3991,
      "mean_token_accuracy": 0.17641332894563674,
      "num_tokens": 13107059.0,
      "step": 7555
    },
    {
      "entropy": 5.6285295486450195,
      "epoch": 0.5882124100369578,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970689330510298,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16372135579586028,
      "num_tokens": 13116783.0,
      "step": 7560
    },
    {
      "entropy": 5.636148834228516,
      "epoch": 0.5886014394086754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970644723229868,
      "loss": 5.3586,
      "mean_token_accuracy": 0.17543469071388246,
      "num_tokens": 13124998.0,
      "step": 7565
    },
    {
      "entropy": 5.6454380512237545,
      "epoch": 0.588990468780393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970600082254687,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1735449880361557,
      "num_tokens": 13134507.0,
      "step": 7570
    },
    {
      "entropy": 5.549423742294311,
      "epoch": 0.5893794981521104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970555407585436,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17249612510204315,
      "num_tokens": 13143682.0,
      "step": 7575
    },
    {
      "entropy": 5.58020977973938,
      "epoch": 0.589768527523828,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497051069922279,
      "loss": 5.4013,
      "mean_token_accuracy": 0.17372599989175797,
      "num_tokens": 13151924.0,
      "step": 7580
    },
    {
      "entropy": 5.5826180934906,
      "epoch": 0.5901575568955456,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970465957167429,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17643746584653855,
      "num_tokens": 13160853.0,
      "step": 7585
    },
    {
      "entropy": 5.621823501586914,
      "epoch": 0.5905465862672632,
      "grad_norm": 1.0,
      "learning_rate": 0.000497042118142003,
      "loss": 5.379,
      "mean_token_accuracy": 0.180886971950531,
      "num_tokens": 13168993.0,
      "step": 7590
    },
    {
      "entropy": 5.562651824951172,
      "epoch": 0.5909356156389808,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970376371981275,
      "loss": 5.3061,
      "mean_token_accuracy": 0.17483243048191072,
      "num_tokens": 13177266.0,
      "step": 7595
    },
    {
      "entropy": 5.621056079864502,
      "epoch": 0.5913246450106983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970331528851842,
      "loss": 5.3839,
      "mean_token_accuracy": 0.16920152455568313,
      "num_tokens": 13185328.0,
      "step": 7600
    },
    {
      "entropy": 5.540424346923828,
      "epoch": 0.5917136743824158,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004970286652032412,
      "loss": 5.2685,
      "mean_token_accuracy": 0.18061549067497254,
      "num_tokens": 13194502.0,
      "step": 7605
    },
    {
      "entropy": 5.563891077041626,
      "epoch": 0.5921027037541334,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970241741523667,
      "loss": 5.4074,
      "mean_token_accuracy": 0.17013546228408813,
      "num_tokens": 13203745.0,
      "step": 7610
    },
    {
      "entropy": 5.606880235671997,
      "epoch": 0.592491733125851,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970196797326286,
      "loss": 5.4611,
      "mean_token_accuracy": 0.1726463630795479,
      "num_tokens": 13213115.0,
      "step": 7615
    },
    {
      "entropy": 5.696786546707154,
      "epoch": 0.5928807624975686,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004970151819440955,
      "loss": 5.484,
      "mean_token_accuracy": 0.167642742395401,
      "num_tokens": 13222231.0,
      "step": 7620
    },
    {
      "entropy": 5.590134191513061,
      "epoch": 0.5932697918692861,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970106807868351,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17260773181915284,
      "num_tokens": 13231346.0,
      "step": 7625
    },
    {
      "entropy": 5.575092601776123,
      "epoch": 0.5936588212410037,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497006176260916,
      "loss": 5.4107,
      "mean_token_accuracy": 0.16619206964969635,
      "num_tokens": 13240436.0,
      "step": 7630
    },
    {
      "entropy": 5.629506921768188,
      "epoch": 0.5940478506127213,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970016683664068,
      "loss": 5.3303,
      "mean_token_accuracy": 0.1771960213780403,
      "num_tokens": 13249501.0,
      "step": 7635
    },
    {
      "entropy": 5.4879992485046385,
      "epoch": 0.5944368799844388,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969971571033754,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18234076499938964,
      "num_tokens": 13258322.0,
      "step": 7640
    },
    {
      "entropy": 5.556401205062866,
      "epoch": 0.5948259093561564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969926424718906,
      "loss": 5.4351,
      "mean_token_accuracy": 0.17190596908330918,
      "num_tokens": 13267418.0,
      "step": 7645
    },
    {
      "entropy": 5.677619123458863,
      "epoch": 0.5952149387278739,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969881244720207,
      "loss": 5.4237,
      "mean_token_accuracy": 0.16786353588104247,
      "num_tokens": 13276243.0,
      "step": 7650
    },
    {
      "entropy": 5.4979438304901125,
      "epoch": 0.5956039680995915,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969836031038343,
      "loss": 5.3041,
      "mean_token_accuracy": 0.16940934658050538,
      "num_tokens": 13285611.0,
      "step": 7655
    },
    {
      "entropy": 5.584447860717773,
      "epoch": 0.5959929974713091,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004969790783674002,
      "loss": 5.3985,
      "mean_token_accuracy": 0.17014468610286712,
      "num_tokens": 13293949.0,
      "step": 7660
    },
    {
      "entropy": 5.576897478103637,
      "epoch": 0.5963820268430267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969745502627868,
      "loss": 5.3419,
      "mean_token_accuracy": 0.17507205307483673,
      "num_tokens": 13302593.0,
      "step": 7665
    },
    {
      "entropy": 5.589267921447754,
      "epoch": 0.5967710562147442,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969700187900629,
      "loss": 5.3911,
      "mean_token_accuracy": 0.169544081389904,
      "num_tokens": 13310947.0,
      "step": 7670
    },
    {
      "entropy": 5.620465183258057,
      "epoch": 0.5971600855864618,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969654839492973,
      "loss": 5.3729,
      "mean_token_accuracy": 0.1709015414118767,
      "num_tokens": 13319233.0,
      "step": 7675
    },
    {
      "entropy": 5.654470491409302,
      "epoch": 0.5975491149581793,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969609457405588,
      "loss": 5.464,
      "mean_token_accuracy": 0.16662180572748184,
      "num_tokens": 13327979.0,
      "step": 7680
    },
    {
      "entropy": 5.604493188858032,
      "epoch": 0.5979381443298969,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969564041639163,
      "loss": 5.4169,
      "mean_token_accuracy": 0.17162067741155623,
      "num_tokens": 13336699.0,
      "step": 7685
    },
    {
      "entropy": 5.585852384567261,
      "epoch": 0.5983271737016145,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969518592194386,
      "loss": 5.299,
      "mean_token_accuracy": 0.1750283420085907,
      "num_tokens": 13344688.0,
      "step": 7690
    },
    {
      "entropy": 5.544575977325439,
      "epoch": 0.5987162030733321,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969473109071946,
      "loss": 5.3237,
      "mean_token_accuracy": 0.17674118876457215,
      "num_tokens": 13353442.0,
      "step": 7695
    },
    {
      "entropy": 5.6312174797058105,
      "epoch": 0.5991052324450497,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969427592272535,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1720339149236679,
      "num_tokens": 13362388.0,
      "step": 7700
    },
    {
      "entropy": 5.483028316497803,
      "epoch": 0.5994942618167671,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004969382041796843,
      "loss": 5.196,
      "mean_token_accuracy": 0.18541455566883086,
      "num_tokens": 13370796.0,
      "step": 7705
    },
    {
      "entropy": 5.510852098464966,
      "epoch": 0.5998832911884847,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496933645764556,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17108747214078904,
      "num_tokens": 13379733.0,
      "step": 7710
    },
    {
      "entropy": 5.534001636505127,
      "epoch": 0.6002723205602023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969290839819381,
      "loss": 5.2606,
      "mean_token_accuracy": 0.1825840100646019,
      "num_tokens": 13388278.0,
      "step": 7715
    },
    {
      "entropy": 5.6160965919494625,
      "epoch": 0.6006613499319199,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969245188318994,
      "loss": 5.4387,
      "mean_token_accuracy": 0.17236709743738174,
      "num_tokens": 13397461.0,
      "step": 7720
    },
    {
      "entropy": 5.53111891746521,
      "epoch": 0.6010503793036375,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969199503145093,
      "loss": 5.307,
      "mean_token_accuracy": 0.1779583439230919,
      "num_tokens": 13406087.0,
      "step": 7725
    },
    {
      "entropy": 5.592005586624145,
      "epoch": 0.601439408675355,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969153784298374,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1804654434323311,
      "num_tokens": 13414760.0,
      "step": 7730
    },
    {
      "entropy": 5.533606243133545,
      "epoch": 0.6018284380470725,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969108031779527,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17098020613193513,
      "num_tokens": 13423574.0,
      "step": 7735
    },
    {
      "entropy": 5.529449605941773,
      "epoch": 0.6022174674187901,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004969062245589247,
      "loss": 5.286,
      "mean_token_accuracy": 0.18063054382801055,
      "num_tokens": 13431799.0,
      "step": 7740
    },
    {
      "entropy": 5.571295166015625,
      "epoch": 0.6026064967905077,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004969016425728231,
      "loss": 5.3392,
      "mean_token_accuracy": 0.17426727712154388,
      "num_tokens": 13440436.0,
      "step": 7745
    },
    {
      "entropy": 5.569225454330445,
      "epoch": 0.6029955261622253,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968970572197172,
      "loss": 5.2949,
      "mean_token_accuracy": 0.18219717293977739,
      "num_tokens": 13449236.0,
      "step": 7750
    },
    {
      "entropy": 5.485515785217285,
      "epoch": 0.6033845555339428,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968924684996765,
      "loss": 5.254,
      "mean_token_accuracy": 0.17624879032373428,
      "num_tokens": 13457697.0,
      "step": 7755
    },
    {
      "entropy": 5.540696716308593,
      "epoch": 0.6037735849056604,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496887876412771,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17585546374320984,
      "num_tokens": 13465939.0,
      "step": 7760
    },
    {
      "entropy": 5.566684436798096,
      "epoch": 0.6041626142773779,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004968832809590701,
      "loss": 5.2813,
      "mean_token_accuracy": 0.18138911575078964,
      "num_tokens": 13474154.0,
      "step": 7765
    },
    {
      "entropy": 5.613201427459717,
      "epoch": 0.6045516436490955,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004968786821386435,
      "loss": 5.2996,
      "mean_token_accuracy": 0.17543721199035645,
      "num_tokens": 13482943.0,
      "step": 7770
    },
    {
      "entropy": 5.597583436965943,
      "epoch": 0.6049406730208131,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004968740799515611,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1688978724181652,
      "num_tokens": 13491945.0,
      "step": 7775
    },
    {
      "entropy": 5.537314510345459,
      "epoch": 0.6053297023925306,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004968694743978926,
      "loss": 5.3675,
      "mean_token_accuracy": 0.16827480494976044,
      "num_tokens": 13501277.0,
      "step": 7780
    },
    {
      "entropy": 5.613169622421265,
      "epoch": 0.6057187317642482,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968648654777081,
      "loss": 5.393,
      "mean_token_accuracy": 0.17645200192928315,
      "num_tokens": 13509609.0,
      "step": 7785
    },
    {
      "entropy": 5.581654214859009,
      "epoch": 0.6061077611359658,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004968602531910773,
      "loss": 5.391,
      "mean_token_accuracy": 0.16612172424793242,
      "num_tokens": 13518707.0,
      "step": 7790
    },
    {
      "entropy": 5.587740039825439,
      "epoch": 0.6064967905076833,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968556375380704,
      "loss": 5.3508,
      "mean_token_accuracy": 0.17670366466045379,
      "num_tokens": 13527973.0,
      "step": 7795
    },
    {
      "entropy": 5.5790623188018795,
      "epoch": 0.6068858198794009,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968510185187571,
      "loss": 5.288,
      "mean_token_accuracy": 0.18264268040657045,
      "num_tokens": 13536580.0,
      "step": 7800
    },
    {
      "entropy": 5.509527015686035,
      "epoch": 0.6072748492511184,
      "grad_norm": 1.09375,
      "learning_rate": 0.000496846396133208,
      "loss": 5.3231,
      "mean_token_accuracy": 0.17557021826505662,
      "num_tokens": 13544826.0,
      "step": 7805
    },
    {
      "entropy": 5.588966751098633,
      "epoch": 0.607663878622836,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004968417703814928,
      "loss": 5.3712,
      "mean_token_accuracy": 0.18416630029678344,
      "num_tokens": 13552821.0,
      "step": 7810
    },
    {
      "entropy": 5.500114679336548,
      "epoch": 0.6080529079945536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968371412636818,
      "loss": 5.2417,
      "mean_token_accuracy": 0.17580069303512574,
      "num_tokens": 13560827.0,
      "step": 7815
    },
    {
      "entropy": 5.547198343276977,
      "epoch": 0.6084419373662712,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004968325087798453,
      "loss": 5.4232,
      "mean_token_accuracy": 0.17011705040931702,
      "num_tokens": 13569311.0,
      "step": 7820
    },
    {
      "entropy": 5.649914121627807,
      "epoch": 0.6088309667379888,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968278729300536,
      "loss": 5.4098,
      "mean_token_accuracy": 0.16785431355237962,
      "num_tokens": 13579035.0,
      "step": 7825
    },
    {
      "entropy": 5.596644639968872,
      "epoch": 0.6092199961097062,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968232337143769,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17886339426040648,
      "num_tokens": 13587508.0,
      "step": 7830
    },
    {
      "entropy": 5.5370063304901125,
      "epoch": 0.6096090254814238,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004968185911328858,
      "loss": 5.4058,
      "mean_token_accuracy": 0.17701992094516755,
      "num_tokens": 13596515.0,
      "step": 7835
    },
    {
      "entropy": 5.544048547744751,
      "epoch": 0.6099980548531414,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968139451856506,
      "loss": 5.2967,
      "mean_token_accuracy": 0.1797937497496605,
      "num_tokens": 13605162.0,
      "step": 7840
    },
    {
      "entropy": 5.561970663070679,
      "epoch": 0.610387084224859,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968092958727419,
      "loss": 5.2798,
      "mean_token_accuracy": 0.17963772565126418,
      "num_tokens": 13614398.0,
      "step": 7845
    },
    {
      "entropy": 5.54109377861023,
      "epoch": 0.6107761135965766,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049680464319423,
      "loss": 5.2511,
      "mean_token_accuracy": 0.17509905099868775,
      "num_tokens": 13622347.0,
      "step": 7850
    },
    {
      "entropy": 5.503065776824951,
      "epoch": 0.6111651429682942,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967999871501858,
      "loss": 5.2657,
      "mean_token_accuracy": 0.17665477693080903,
      "num_tokens": 13630773.0,
      "step": 7855
    },
    {
      "entropy": 5.551439714431763,
      "epoch": 0.6115541723400116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967953277406798,
      "loss": 5.2635,
      "mean_token_accuracy": 0.1790945053100586,
      "num_tokens": 13639857.0,
      "step": 7860
    },
    {
      "entropy": 5.6004712104797365,
      "epoch": 0.6119432017117292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967906649657828,
      "loss": 5.3727,
      "mean_token_accuracy": 0.17795006930828094,
      "num_tokens": 13649959.0,
      "step": 7865
    },
    {
      "entropy": 5.625068473815918,
      "epoch": 0.6123322310834468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004967859988255655,
      "loss": 5.3995,
      "mean_token_accuracy": 0.17279022485017775,
      "num_tokens": 13658072.0,
      "step": 7870
    },
    {
      "entropy": 5.529886484146118,
      "epoch": 0.6127212604551644,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967813293200985,
      "loss": 5.3382,
      "mean_token_accuracy": 0.18108959943056108,
      "num_tokens": 13666816.0,
      "step": 7875
    },
    {
      "entropy": 5.525184059143067,
      "epoch": 0.613110289826882,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004967766564494529,
      "loss": 5.1971,
      "mean_token_accuracy": 0.18741032481193542,
      "num_tokens": 13675415.0,
      "step": 7880
    },
    {
      "entropy": 5.510573244094848,
      "epoch": 0.6134993191985995,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004967719802136996,
      "loss": 5.2432,
      "mean_token_accuracy": 0.18263396471738816,
      "num_tokens": 13684090.0,
      "step": 7885
    },
    {
      "entropy": 5.516405916213989,
      "epoch": 0.613888348570317,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004967673006129094,
      "loss": 5.2915,
      "mean_token_accuracy": 0.18200115412473677,
      "num_tokens": 13692515.0,
      "step": 7890
    },
    {
      "entropy": 5.516071128845215,
      "epoch": 0.6142773779420346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967626176471536,
      "loss": 5.356,
      "mean_token_accuracy": 0.1775418221950531,
      "num_tokens": 13700780.0,
      "step": 7895
    },
    {
      "entropy": 5.636789226531983,
      "epoch": 0.6146664073137522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967579313165028,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1797405645251274,
      "num_tokens": 13708638.0,
      "step": 7900
    },
    {
      "entropy": 5.601937341690063,
      "epoch": 0.6150554366854698,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004967532416210284,
      "loss": 5.3831,
      "mean_token_accuracy": 0.1653753012418747,
      "num_tokens": 13717867.0,
      "step": 7905
    },
    {
      "entropy": 5.563804340362549,
      "epoch": 0.6154444660571873,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967485485608016,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17676510214805602,
      "num_tokens": 13727280.0,
      "step": 7910
    },
    {
      "entropy": 5.595907878875733,
      "epoch": 0.6158334954289049,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967438521358934,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17540148198604583,
      "num_tokens": 13735255.0,
      "step": 7915
    },
    {
      "entropy": 5.532194900512695,
      "epoch": 0.6162225248006225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967391523463754,
      "loss": 5.3775,
      "mean_token_accuracy": 0.17733073532581328,
      "num_tokens": 13743418.0,
      "step": 7920
    },
    {
      "entropy": 5.543993854522705,
      "epoch": 0.61661155417234,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967344491923185,
      "loss": 5.2762,
      "mean_token_accuracy": 0.17798438519239426,
      "num_tokens": 13752714.0,
      "step": 7925
    },
    {
      "entropy": 5.6618695735931395,
      "epoch": 0.6170005835440576,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004967297426737943,
      "loss": 5.4486,
      "mean_token_accuracy": 0.16520682275295256,
      "num_tokens": 13761350.0,
      "step": 7930
    },
    {
      "entropy": 5.5297904968261715,
      "epoch": 0.6173896129157751,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004967250327908742,
      "loss": 5.2824,
      "mean_token_accuracy": 0.17875937521457672,
      "num_tokens": 13769769.0,
      "step": 7935
    },
    {
      "entropy": 5.550059127807617,
      "epoch": 0.6177786422874927,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004967203195436297,
      "loss": 5.311,
      "mean_token_accuracy": 0.17310796827077865,
      "num_tokens": 13778432.0,
      "step": 7940
    },
    {
      "entropy": 5.5211821556091305,
      "epoch": 0.6181676716592103,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004967156029321322,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17745684683322907,
      "num_tokens": 13786968.0,
      "step": 7945
    },
    {
      "entropy": 5.61552062034607,
      "epoch": 0.6185567010309279,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967108829564532,
      "loss": 5.4223,
      "mean_token_accuracy": 0.16451818495988846,
      "num_tokens": 13795824.0,
      "step": 7950
    },
    {
      "entropy": 5.629518461227417,
      "epoch": 0.6189457304026454,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004967061596166646,
      "loss": 5.3494,
      "mean_token_accuracy": 0.17811042666435242,
      "num_tokens": 13804748.0,
      "step": 7955
    },
    {
      "entropy": 5.509480619430542,
      "epoch": 0.6193347597743629,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004967014329128378,
      "loss": 5.3398,
      "mean_token_accuracy": 0.17735178172588348,
      "num_tokens": 13813106.0,
      "step": 7960
    },
    {
      "entropy": 5.555459022521973,
      "epoch": 0.6197237891460805,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966967028450448,
      "loss": 5.3667,
      "mean_token_accuracy": 0.17630673050880433,
      "num_tokens": 13821787.0,
      "step": 7965
    },
    {
      "entropy": 5.603826332092285,
      "epoch": 0.6201128185177981,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496691969413357,
      "loss": 5.3539,
      "mean_token_accuracy": 0.17787329405546187,
      "num_tokens": 13830275.0,
      "step": 7970
    },
    {
      "entropy": 5.557550859451294,
      "epoch": 0.6205018478895157,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966872326178466,
      "loss": 5.2798,
      "mean_token_accuracy": 0.1828768104314804,
      "num_tokens": 13838868.0,
      "step": 7975
    },
    {
      "entropy": 5.551296949386597,
      "epoch": 0.6208908772612333,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966824924585851,
      "loss": 5.3555,
      "mean_token_accuracy": 0.17854246348142624,
      "num_tokens": 13846970.0,
      "step": 7980
    },
    {
      "entropy": 5.613220357894898,
      "epoch": 0.6212799066329507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966777489356448,
      "loss": 5.4196,
      "mean_token_accuracy": 0.17233560085296631,
      "num_tokens": 13856059.0,
      "step": 7985
    },
    {
      "entropy": 5.587001371383667,
      "epoch": 0.6216689360046683,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966730020490975,
      "loss": 5.3059,
      "mean_token_accuracy": 0.17722841799259187,
      "num_tokens": 13865461.0,
      "step": 7990
    },
    {
      "entropy": 5.5760759830474855,
      "epoch": 0.6220579653763859,
      "grad_norm": 0.96875,
      "learning_rate": 0.000496668251799015,
      "loss": 5.2765,
      "mean_token_accuracy": 0.17371441125869752,
      "num_tokens": 13874124.0,
      "step": 7995
    },
    {
      "entropy": 5.613064813613891,
      "epoch": 0.6224469947481035,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004966634981854699,
      "loss": 5.3814,
      "mean_token_accuracy": 0.17237360179424285,
      "num_tokens": 13882952.0,
      "step": 8000
    },
    {
      "entropy": 5.585406970977783,
      "epoch": 0.6228360241198211,
      "grad_norm": 1.0,
      "learning_rate": 0.0004966587412085339,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17545945644378663,
      "num_tokens": 13890687.0,
      "step": 8005
    },
    {
      "entropy": 5.577899980545044,
      "epoch": 0.6232250534915386,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004966539808682793,
      "loss": 5.4074,
      "mean_token_accuracy": 0.1689345955848694,
      "num_tokens": 13899773.0,
      "step": 8010
    },
    {
      "entropy": 5.468811941146851,
      "epoch": 0.6236140828632561,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004966492171647783,
      "loss": 5.1914,
      "mean_token_accuracy": 0.18073903769254684,
      "num_tokens": 13908313.0,
      "step": 8015
    },
    {
      "entropy": 5.554126024246216,
      "epoch": 0.6240031122349737,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004966444500981032,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18334477245807648,
      "num_tokens": 13916422.0,
      "step": 8020
    },
    {
      "entropy": 5.580251407623291,
      "epoch": 0.6243921416066913,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966396796683265,
      "loss": 5.3621,
      "mean_token_accuracy": 0.17060983330011367,
      "num_tokens": 13924971.0,
      "step": 8025
    },
    {
      "entropy": 5.572302627563476,
      "epoch": 0.6247811709784089,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966349058755204,
      "loss": 5.2716,
      "mean_token_accuracy": 0.17709621489048005,
      "num_tokens": 13933016.0,
      "step": 8030
    },
    {
      "entropy": 5.673058271408081,
      "epoch": 0.6251702003501264,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004966301287197573,
      "loss": 5.3638,
      "mean_token_accuracy": 0.17842250913381577,
      "num_tokens": 13941814.0,
      "step": 8035
    },
    {
      "entropy": 5.531317281723022,
      "epoch": 0.625559229721844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966253482011098,
      "loss": 5.1933,
      "mean_token_accuracy": 0.1801487997174263,
      "num_tokens": 13950344.0,
      "step": 8040
    },
    {
      "entropy": 5.555147981643676,
      "epoch": 0.6259482590935616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966205643196503,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1682429000735283,
      "num_tokens": 13959456.0,
      "step": 8045
    },
    {
      "entropy": 5.592694234848023,
      "epoch": 0.6263372884652791,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966157770754516,
      "loss": 5.3065,
      "mean_token_accuracy": 0.18245015740394593,
      "num_tokens": 13967303.0,
      "step": 8050
    },
    {
      "entropy": 5.585998249053955,
      "epoch": 0.6267263178369967,
      "grad_norm": 0.96875,
      "learning_rate": 0.000496610986468586,
      "loss": 5.3509,
      "mean_token_accuracy": 0.16850290894508363,
      "num_tokens": 13977138.0,
      "step": 8055
    },
    {
      "entropy": 5.638797283172607,
      "epoch": 0.6271153472087143,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004966061924991267,
      "loss": 5.4227,
      "mean_token_accuracy": 0.17261697947978974,
      "num_tokens": 13985062.0,
      "step": 8060
    },
    {
      "entropy": 5.668380260467529,
      "epoch": 0.6275043765804318,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496601395167146,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1716477170586586,
      "num_tokens": 13992923.0,
      "step": 8065
    },
    {
      "entropy": 5.543324136734009,
      "epoch": 0.6278934059521494,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965965944727168,
      "loss": 5.4132,
      "mean_token_accuracy": 0.16822911947965621,
      "num_tokens": 14002116.0,
      "step": 8070
    },
    {
      "entropy": 5.580185413360596,
      "epoch": 0.628282435323867,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004965917904159121,
      "loss": 5.3931,
      "mean_token_accuracy": 0.17099771350622178,
      "num_tokens": 14011053.0,
      "step": 8075
    },
    {
      "entropy": 5.630213594436645,
      "epoch": 0.6286714646955845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965869829968046,
      "loss": 5.3342,
      "mean_token_accuracy": 0.1766078531742096,
      "num_tokens": 14019226.0,
      "step": 8080
    },
    {
      "entropy": 5.4770303726196286,
      "epoch": 0.6290604940673021,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004965821722154674,
      "loss": 5.3391,
      "mean_token_accuracy": 0.17408648133277893,
      "num_tokens": 14027630.0,
      "step": 8085
    },
    {
      "entropy": 5.544252061843872,
      "epoch": 0.6294495234390196,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965773580719734,
      "loss": 5.2561,
      "mean_token_accuracy": 0.18546278923749923,
      "num_tokens": 14036004.0,
      "step": 8090
    },
    {
      "entropy": 5.5833745956420895,
      "epoch": 0.6298385528107372,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965725405663957,
      "loss": 5.3554,
      "mean_token_accuracy": 0.17322381734848022,
      "num_tokens": 14045359.0,
      "step": 8095
    },
    {
      "entropy": 5.542785596847534,
      "epoch": 0.6302275821824548,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004965677196988073,
      "loss": 5.2898,
      "mean_token_accuracy": 0.1769864469766617,
      "num_tokens": 14054099.0,
      "step": 8100
    },
    {
      "entropy": 5.564396667480469,
      "epoch": 0.6306166115541724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965628954692815,
      "loss": 5.3129,
      "mean_token_accuracy": 0.18053655177354813,
      "num_tokens": 14062414.0,
      "step": 8105
    },
    {
      "entropy": 5.538123321533203,
      "epoch": 0.63100564092589,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004965580678778913,
      "loss": 5.3513,
      "mean_token_accuracy": 0.17966470420360564,
      "num_tokens": 14071016.0,
      "step": 8110
    },
    {
      "entropy": 5.558867883682251,
      "epoch": 0.6313946702976074,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965532369247102,
      "loss": 5.343,
      "mean_token_accuracy": 0.17081669420003892,
      "num_tokens": 14080622.0,
      "step": 8115
    },
    {
      "entropy": 5.537043571472168,
      "epoch": 0.631783699669325,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965484026098111,
      "loss": 5.1584,
      "mean_token_accuracy": 0.18677742630243302,
      "num_tokens": 14089500.0,
      "step": 8120
    },
    {
      "entropy": 5.58234395980835,
      "epoch": 0.6321727290410426,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004965435649332679,
      "loss": 5.4148,
      "mean_token_accuracy": 0.17902248948812485,
      "num_tokens": 14098260.0,
      "step": 8125
    },
    {
      "entropy": 5.554372978210449,
      "epoch": 0.6325617584127602,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004965387238951536,
      "loss": 5.2376,
      "mean_token_accuracy": 0.1855595663189888,
      "num_tokens": 14106561.0,
      "step": 8130
    },
    {
      "entropy": 5.603929567337036,
      "epoch": 0.6329507877844778,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965338794955418,
      "loss": 5.338,
      "mean_token_accuracy": 0.1756177231669426,
      "num_tokens": 14115049.0,
      "step": 8135
    },
    {
      "entropy": 5.541755533218383,
      "epoch": 0.6333398171561953,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965290317345061,
      "loss": 5.2387,
      "mean_token_accuracy": 0.17843602746725082,
      "num_tokens": 14123204.0,
      "step": 8140
    },
    {
      "entropy": 5.4808355331420895,
      "epoch": 0.6337288465279128,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004965241806121198,
      "loss": 5.2868,
      "mean_token_accuracy": 0.18215357661247253,
      "num_tokens": 14132146.0,
      "step": 8145
    },
    {
      "entropy": 5.628285121917725,
      "epoch": 0.6341178758996304,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004965193261284567,
      "loss": 5.3899,
      "mean_token_accuracy": 0.16875875890254974,
      "num_tokens": 14140839.0,
      "step": 8150
    },
    {
      "entropy": 5.610815334320068,
      "epoch": 0.634506905271348,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965144682835904,
      "loss": 5.3795,
      "mean_token_accuracy": 0.1781456559896469,
      "num_tokens": 14149551.0,
      "step": 8155
    },
    {
      "entropy": 5.515106868743897,
      "epoch": 0.6348959346430656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965096070775946,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1755740150809288,
      "num_tokens": 14158414.0,
      "step": 8160
    },
    {
      "entropy": 5.577139282226563,
      "epoch": 0.6352849640147831,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004965047425105431,
      "loss": 5.3882,
      "mean_token_accuracy": 0.17380691021680833,
      "num_tokens": 14167984.0,
      "step": 8165
    },
    {
      "entropy": 5.535385370254517,
      "epoch": 0.6356739933865007,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004964998745825097,
      "loss": 5.316,
      "mean_token_accuracy": 0.17888084948062896,
      "num_tokens": 14177244.0,
      "step": 8170
    },
    {
      "entropy": 5.588834619522094,
      "epoch": 0.6360630227582182,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964950032935682,
      "loss": 5.2512,
      "mean_token_accuracy": 0.1779651641845703,
      "num_tokens": 14185754.0,
      "step": 8175
    },
    {
      "entropy": 5.528890562057495,
      "epoch": 0.6364520521299358,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964901286437927,
      "loss": 5.2899,
      "mean_token_accuracy": 0.1799963116645813,
      "num_tokens": 14194242.0,
      "step": 8180
    },
    {
      "entropy": 5.524556064605713,
      "epoch": 0.6368410815016534,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964852506332568,
      "loss": 5.2153,
      "mean_token_accuracy": 0.18313002586364746,
      "num_tokens": 14202496.0,
      "step": 8185
    },
    {
      "entropy": 5.5839591979980465,
      "epoch": 0.6372301108733709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496480369262035,
      "loss": 5.448,
      "mean_token_accuracy": 0.17073566168546678,
      "num_tokens": 14210510.0,
      "step": 8190
    },
    {
      "entropy": 5.605168581008911,
      "epoch": 0.6376191402450885,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496475484530201,
      "loss": 5.4347,
      "mean_token_accuracy": 0.16247031688690186,
      "num_tokens": 14219238.0,
      "step": 8195
    },
    {
      "entropy": 5.502722358703613,
      "epoch": 0.6380081696168061,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964705964378292,
      "loss": 5.243,
      "mean_token_accuracy": 0.18735675364732743,
      "num_tokens": 14227651.0,
      "step": 8200
    },
    {
      "entropy": 5.496855878829956,
      "epoch": 0.6383971989885237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004964657049849934,
      "loss": 5.2097,
      "mean_token_accuracy": 0.18215781450271606,
      "num_tokens": 14236144.0,
      "step": 8205
    },
    {
      "entropy": 5.635563611984253,
      "epoch": 0.6387862283602412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004964608101717681,
      "loss": 5.3734,
      "mean_token_accuracy": 0.16718583106994628,
      "num_tokens": 14244706.0,
      "step": 8210
    },
    {
      "entropy": 5.574279880523681,
      "epoch": 0.6391752577319587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964559119982275,
      "loss": 5.3199,
      "mean_token_accuracy": 0.18086613714694977,
      "num_tokens": 14252873.0,
      "step": 8215
    },
    {
      "entropy": 5.576375818252563,
      "epoch": 0.6395642871036763,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496451010464446,
      "loss": 5.3899,
      "mean_token_accuracy": 0.1750391438603401,
      "num_tokens": 14261383.0,
      "step": 8220
    },
    {
      "entropy": 5.475570201873779,
      "epoch": 0.6399533164753939,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004964461055704978,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18437148928642272,
      "num_tokens": 14270087.0,
      "step": 8225
    },
    {
      "entropy": 5.6076884269714355,
      "epoch": 0.6403423458471115,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004964411973164574,
      "loss": 5.3609,
      "mean_token_accuracy": 0.18397365808486937,
      "num_tokens": 14278826.0,
      "step": 8230
    },
    {
      "entropy": 5.507570075988769,
      "epoch": 0.6407313752188291,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964362857023993,
      "loss": 5.2796,
      "mean_token_accuracy": 0.17902056723833085,
      "num_tokens": 14287554.0,
      "step": 8235
    },
    {
      "entropy": 5.566535139083863,
      "epoch": 0.6411204045905465,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000496431370728398,
      "loss": 5.2667,
      "mean_token_accuracy": 0.17709399312734603,
      "num_tokens": 14295839.0,
      "step": 8240
    },
    {
      "entropy": 5.555580949783325,
      "epoch": 0.6415094339622641,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964264523945281,
      "loss": 5.3024,
      "mean_token_accuracy": 0.1806720845401287,
      "num_tokens": 14304831.0,
      "step": 8245
    },
    {
      "entropy": 5.575355482101441,
      "epoch": 0.6418984633339817,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964215307008643,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17711037695407866,
      "num_tokens": 14313249.0,
      "step": 8250
    },
    {
      "entropy": 5.573872232437134,
      "epoch": 0.6422874927056993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004964166056474811,
      "loss": 5.3412,
      "mean_token_accuracy": 0.17384667545557023,
      "num_tokens": 14321978.0,
      "step": 8255
    },
    {
      "entropy": 5.4906227588653564,
      "epoch": 0.6426765220774169,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964116772344534,
      "loss": 5.2501,
      "mean_token_accuracy": 0.17830630242824555,
      "num_tokens": 14330379.0,
      "step": 8260
    },
    {
      "entropy": 5.5215977191925045,
      "epoch": 0.6430655514491345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964067454618559,
      "loss": 5.2935,
      "mean_token_accuracy": 0.18217869699001313,
      "num_tokens": 14339121.0,
      "step": 8265
    },
    {
      "entropy": 5.504442310333252,
      "epoch": 0.6434545808208519,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964018103297634,
      "loss": 5.247,
      "mean_token_accuracy": 0.18260000348091127,
      "num_tokens": 14347443.0,
      "step": 8270
    },
    {
      "entropy": 5.5835634708404545,
      "epoch": 0.6438436101925695,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004963968718382509,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17404672801494597,
      "num_tokens": 14356459.0,
      "step": 8275
    },
    {
      "entropy": 5.550328874588013,
      "epoch": 0.6442326395642871,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004963919299873931,
      "loss": 5.2589,
      "mean_token_accuracy": 0.17677343338727952,
      "num_tokens": 14365359.0,
      "step": 8280
    },
    {
      "entropy": 5.450421857833862,
      "epoch": 0.6446216689360047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963869847772653,
      "loss": 5.196,
      "mean_token_accuracy": 0.18321047723293304,
      "num_tokens": 14373345.0,
      "step": 8285
    },
    {
      "entropy": 5.513602542877197,
      "epoch": 0.6450106983077223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963820362079423,
      "loss": 5.2799,
      "mean_token_accuracy": 0.18093516677618027,
      "num_tokens": 14381977.0,
      "step": 8290
    },
    {
      "entropy": 5.462281084060669,
      "epoch": 0.6453997276794398,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004963770842794994,
      "loss": 5.1796,
      "mean_token_accuracy": 0.1848599746823311,
      "num_tokens": 14390486.0,
      "step": 8295
    },
    {
      "entropy": 5.5589361667633055,
      "epoch": 0.6457887570511573,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963721289920115,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17912714034318925,
      "num_tokens": 14399064.0,
      "step": 8300
    },
    {
      "entropy": 5.539777040481567,
      "epoch": 0.6461777864228749,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963671703455538,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1839711531996727,
      "num_tokens": 14406965.0,
      "step": 8305
    },
    {
      "entropy": 5.482750988006591,
      "epoch": 0.6465668157945925,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004963622083402018,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17665068954229354,
      "num_tokens": 14416293.0,
      "step": 8310
    },
    {
      "entropy": 5.606808519363403,
      "epoch": 0.6469558451663101,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963572429760305,
      "loss": 5.3405,
      "mean_token_accuracy": 0.17011832147836686,
      "num_tokens": 14424515.0,
      "step": 8315
    },
    {
      "entropy": 5.556892204284668,
      "epoch": 0.6473448745380276,
      "grad_norm": 1.0,
      "learning_rate": 0.0004963522742531155,
      "loss": 5.1785,
      "mean_token_accuracy": 0.18246434181928634,
      "num_tokens": 14432706.0,
      "step": 8320
    },
    {
      "entropy": 5.480896806716919,
      "epoch": 0.6477339039097452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496347302171532,
      "loss": 5.2834,
      "mean_token_accuracy": 0.1854716181755066,
      "num_tokens": 14441462.0,
      "step": 8325
    },
    {
      "entropy": 5.467887020111084,
      "epoch": 0.6481229332814628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963423267313554,
      "loss": 5.2568,
      "mean_token_accuracy": 0.1838444337248802,
      "num_tokens": 14450218.0,
      "step": 8330
    },
    {
      "entropy": 5.616094160079956,
      "epoch": 0.6485119626531803,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004963373479326614,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16786930859088897,
      "num_tokens": 14459825.0,
      "step": 8335
    },
    {
      "entropy": 5.570043230056763,
      "epoch": 0.6489009920248979,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004963323657755255,
      "loss": 5.2699,
      "mean_token_accuracy": 0.17940721809864044,
      "num_tokens": 14468647.0,
      "step": 8340
    },
    {
      "entropy": 5.465993928909302,
      "epoch": 0.6492900213966154,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963273802600231,
      "loss": 5.1626,
      "mean_token_accuracy": 0.1850014716386795,
      "num_tokens": 14476892.0,
      "step": 8345
    },
    {
      "entropy": 5.603579473495484,
      "epoch": 0.649679050768333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049632239138623,
      "loss": 5.429,
      "mean_token_accuracy": 0.17392351776361464,
      "num_tokens": 14485541.0,
      "step": 8350
    },
    {
      "entropy": 5.520974779129029,
      "epoch": 0.6500680801400506,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004963173991542219,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18615519255399704,
      "num_tokens": 14494184.0,
      "step": 8355
    },
    {
      "entropy": 5.593913173675537,
      "epoch": 0.6504571095117682,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004963124035640746,
      "loss": 5.3781,
      "mean_token_accuracy": 0.1692001700401306,
      "num_tokens": 14502151.0,
      "step": 8360
    },
    {
      "entropy": 5.55236496925354,
      "epoch": 0.6508461388834857,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963074046158637,
      "loss": 5.3677,
      "mean_token_accuracy": 0.18047490417957307,
      "num_tokens": 14511254.0,
      "step": 8365
    },
    {
      "entropy": 5.6321656703948975,
      "epoch": 0.6512351682552032,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963024023096652,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16682166159152984,
      "num_tokens": 14520492.0,
      "step": 8370
    },
    {
      "entropy": 5.631375026702881,
      "epoch": 0.6516241976269208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962973966455551,
      "loss": 5.3601,
      "mean_token_accuracy": 0.16875505596399307,
      "num_tokens": 14530082.0,
      "step": 8375
    },
    {
      "entropy": 5.546644496917724,
      "epoch": 0.6520132269986384,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004962923876236092,
      "loss": 5.3218,
      "mean_token_accuracy": 0.17745062112808227,
      "num_tokens": 14539124.0,
      "step": 8380
    },
    {
      "entropy": 5.607315874099731,
      "epoch": 0.652402256370356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004962873752439035,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1814492389559746,
      "num_tokens": 14547780.0,
      "step": 8385
    },
    {
      "entropy": 5.554334831237793,
      "epoch": 0.6527912857420736,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004962823595065141,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1786617323756218,
      "num_tokens": 14556225.0,
      "step": 8390
    },
    {
      "entropy": 5.594890832901001,
      "epoch": 0.653180315113791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004962773404115172,
      "loss": 5.3183,
      "mean_token_accuracy": 0.1761225014925003,
      "num_tokens": 14564147.0,
      "step": 8395
    },
    {
      "entropy": 5.551420640945435,
      "epoch": 0.6535693444855086,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004962723179589886,
      "loss": 5.3068,
      "mean_token_accuracy": 0.17587434202432634,
      "num_tokens": 14573519.0,
      "step": 8400
    },
    {
      "entropy": 5.59135627746582,
      "epoch": 0.6539583738572262,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496267292149005,
      "loss": 5.4309,
      "mean_token_accuracy": 0.16442757546901704,
      "num_tokens": 14582806.0,
      "step": 8405
    },
    {
      "entropy": 5.543904972076416,
      "epoch": 0.6543474032289438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962622629816423,
      "loss": 5.3068,
      "mean_token_accuracy": 0.1739957958459854,
      "num_tokens": 14591274.0,
      "step": 8410
    },
    {
      "entropy": 5.530463457107544,
      "epoch": 0.6547364326006614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000496257230456977,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17313478589057923,
      "num_tokens": 14599902.0,
      "step": 8415
    },
    {
      "entropy": 5.493044996261597,
      "epoch": 0.6551254619723789,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004962521945750855,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17716848850250244,
      "num_tokens": 14608931.0,
      "step": 8420
    },
    {
      "entropy": 5.530213308334351,
      "epoch": 0.6555144913440965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496247155336044,
      "loss": 5.2194,
      "mean_token_accuracy": 0.18342820256948472,
      "num_tokens": 14617547.0,
      "step": 8425
    },
    {
      "entropy": 5.550770044326782,
      "epoch": 0.655903520715814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004962421127399291,
      "loss": 5.3447,
      "mean_token_accuracy": 0.17101119011640548,
      "num_tokens": 14626523.0,
      "step": 8430
    },
    {
      "entropy": 5.586200857162476,
      "epoch": 0.6562925500875316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004962370667868172,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1851767346262932,
      "num_tokens": 14634846.0,
      "step": 8435
    },
    {
      "entropy": 5.528326034545898,
      "epoch": 0.6566815794592492,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962320174767851,
      "loss": 5.294,
      "mean_token_accuracy": 0.1716813012957573,
      "num_tokens": 14643092.0,
      "step": 8440
    },
    {
      "entropy": 5.591558742523193,
      "epoch": 0.6570706088309667,
      "grad_norm": 0.96875,
      "learning_rate": 0.000496226964809909,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1788558319211006,
      "num_tokens": 14651456.0,
      "step": 8445
    },
    {
      "entropy": 5.539442729949951,
      "epoch": 0.6574596382026843,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496221908786266,
      "loss": 5.2565,
      "mean_token_accuracy": 0.18235833197832108,
      "num_tokens": 14660488.0,
      "step": 8450
    },
    {
      "entropy": 5.618861961364746,
      "epoch": 0.6578486675744019,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004962168494059327,
      "loss": 5.4666,
      "mean_token_accuracy": 0.17066026628017425,
      "num_tokens": 14670131.0,
      "step": 8455
    },
    {
      "entropy": 5.613412618637085,
      "epoch": 0.6582376969461194,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962117866689857,
      "loss": 5.2624,
      "mean_token_accuracy": 0.1764333188533783,
      "num_tokens": 14678402.0,
      "step": 8460
    },
    {
      "entropy": 5.541571140289307,
      "epoch": 0.658626726317837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496206720575502,
      "loss": 5.2897,
      "mean_token_accuracy": 0.16646923422813414,
      "num_tokens": 14687589.0,
      "step": 8465
    },
    {
      "entropy": 5.584652948379516,
      "epoch": 0.6590157556895546,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962016511255584,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1722184956073761,
      "num_tokens": 14696288.0,
      "step": 8470
    },
    {
      "entropy": 5.634091234207153,
      "epoch": 0.6594047850612721,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961965783192318,
      "loss": 5.3419,
      "mean_token_accuracy": 0.17304196506738662,
      "num_tokens": 14705269.0,
      "step": 8475
    },
    {
      "entropy": 5.558816766738891,
      "epoch": 0.6597938144329897,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004961915021565992,
      "loss": 5.3162,
      "mean_token_accuracy": 0.17708727568387986,
      "num_tokens": 14713669.0,
      "step": 8480
    },
    {
      "entropy": 5.500200891494751,
      "epoch": 0.6601828438047073,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004961864226377377,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17835226505994797,
      "num_tokens": 14722529.0,
      "step": 8485
    },
    {
      "entropy": 5.575636339187622,
      "epoch": 0.6605718731764249,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004961813397627241,
      "loss": 5.3458,
      "mean_token_accuracy": 0.18031770139932632,
      "num_tokens": 14730956.0,
      "step": 8490
    },
    {
      "entropy": 5.5568760395050045,
      "epoch": 0.6609609025481424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961762535316358,
      "loss": 5.3101,
      "mean_token_accuracy": 0.18019221276044844,
      "num_tokens": 14739320.0,
      "step": 8495
    },
    {
      "entropy": 5.523578739166259,
      "epoch": 0.6613499319198599,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961711639445499,
      "loss": 5.3665,
      "mean_token_accuracy": 0.17696610540151597,
      "num_tokens": 14748030.0,
      "step": 8500
    },
    {
      "entropy": 5.552655601501465,
      "epoch": 0.6617389612915775,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004961660710015435,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18410542160272597,
      "num_tokens": 14756786.0,
      "step": 8505
    },
    {
      "entropy": 5.567683458328247,
      "epoch": 0.6621279906632951,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961609747026943,
      "loss": 5.3213,
      "mean_token_accuracy": 0.1770300105214119,
      "num_tokens": 14765538.0,
      "step": 8510
    },
    {
      "entropy": 5.5553281784057615,
      "epoch": 0.6625170200350127,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961558750480791,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18100422173738479,
      "num_tokens": 14773845.0,
      "step": 8515
    },
    {
      "entropy": 5.499459505081177,
      "epoch": 0.6629060494067303,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961507720377756,
      "loss": 5.2031,
      "mean_token_accuracy": 0.18550216108560563,
      "num_tokens": 14782022.0,
      "step": 8520
    },
    {
      "entropy": 5.564083671569824,
      "epoch": 0.6632950787784477,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961456656718611,
      "loss": 5.343,
      "mean_token_accuracy": 0.18136597871780397,
      "num_tokens": 14790150.0,
      "step": 8525
    },
    {
      "entropy": 5.549467277526856,
      "epoch": 0.6636841081501653,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000496140555950413,
      "loss": 5.379,
      "mean_token_accuracy": 0.17202800512313843,
      "num_tokens": 14799239.0,
      "step": 8530
    },
    {
      "entropy": 5.526183414459228,
      "epoch": 0.6640731375218829,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004961354428735091,
      "loss": 5.3029,
      "mean_token_accuracy": 0.16849559545516968,
      "num_tokens": 14808740.0,
      "step": 8535
    },
    {
      "entropy": 5.681412315368652,
      "epoch": 0.6644621668936005,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004961303264412267,
      "loss": 5.4267,
      "mean_token_accuracy": 0.17152373641729354,
      "num_tokens": 14817329.0,
      "step": 8540
    },
    {
      "entropy": 5.554582118988037,
      "epoch": 0.6648511962653181,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961252066536437,
      "loss": 5.279,
      "mean_token_accuracy": 0.17454319149255754,
      "num_tokens": 14826240.0,
      "step": 8545
    },
    {
      "entropy": 5.548708868026734,
      "epoch": 0.6652402256370356,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961200835108375,
      "loss": 5.299,
      "mean_token_accuracy": 0.1745982825756073,
      "num_tokens": 14834471.0,
      "step": 8550
    },
    {
      "entropy": 5.569959831237793,
      "epoch": 0.6656292550087531,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496114957012886,
      "loss": 5.304,
      "mean_token_accuracy": 0.16618645787239075,
      "num_tokens": 14842724.0,
      "step": 8555
    },
    {
      "entropy": 5.517120409011841,
      "epoch": 0.6660182843804707,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004961098271598671,
      "loss": 5.218,
      "mean_token_accuracy": 0.18385373651981354,
      "num_tokens": 14850725.0,
      "step": 8560
    },
    {
      "entropy": 5.458403539657593,
      "epoch": 0.6664073137521883,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961046939518585,
      "loss": 5.19,
      "mean_token_accuracy": 0.18690327554941177,
      "num_tokens": 14859412.0,
      "step": 8565
    },
    {
      "entropy": 5.5778275489807125,
      "epoch": 0.6667963431239059,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004960995573889379,
      "loss": 5.2877,
      "mean_token_accuracy": 0.17646803557872773,
      "num_tokens": 14867526.0,
      "step": 8570
    },
    {
      "entropy": 5.488861894607544,
      "epoch": 0.6671853724956234,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960944174711836,
      "loss": 5.1961,
      "mean_token_accuracy": 0.17473305016756058,
      "num_tokens": 14875770.0,
      "step": 8575
    },
    {
      "entropy": 5.506997108459473,
      "epoch": 0.667574401867341,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960892741986734,
      "loss": 5.2818,
      "mean_token_accuracy": 0.18370694667100906,
      "num_tokens": 14883872.0,
      "step": 8580
    },
    {
      "entropy": 5.532104063034057,
      "epoch": 0.6679634312390585,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960841275714853,
      "loss": 5.2978,
      "mean_token_accuracy": 0.17296381145715714,
      "num_tokens": 14891526.0,
      "step": 8585
    },
    {
      "entropy": 5.5480423927307125,
      "epoch": 0.6683524606107761,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004960789775896975,
      "loss": 5.3519,
      "mean_token_accuracy": 0.17991099506616592,
      "num_tokens": 14900171.0,
      "step": 8590
    },
    {
      "entropy": 5.594220685958862,
      "epoch": 0.6687414899824937,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960738242533881,
      "loss": 5.3147,
      "mean_token_accuracy": 0.1777023196220398,
      "num_tokens": 14908537.0,
      "step": 8595
    },
    {
      "entropy": 5.508582925796508,
      "epoch": 0.6691305193542112,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004960686675626353,
      "loss": 5.2544,
      "mean_token_accuracy": 0.17995063066482545,
      "num_tokens": 14917387.0,
      "step": 8600
    },
    {
      "entropy": 5.5501314163208,
      "epoch": 0.6695195487259288,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960635075175173,
      "loss": 5.276,
      "mean_token_accuracy": 0.17990948408842086,
      "num_tokens": 14925945.0,
      "step": 8605
    },
    {
      "entropy": 5.539242887496949,
      "epoch": 0.6699085780976464,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960583441181124,
      "loss": 5.2811,
      "mean_token_accuracy": 0.17707243263721467,
      "num_tokens": 14934002.0,
      "step": 8610
    },
    {
      "entropy": 5.532008934020996,
      "epoch": 0.670297607469364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960531773644992,
      "loss": 5.316,
      "mean_token_accuracy": 0.16889427751302719,
      "num_tokens": 14942658.0,
      "step": 8615
    },
    {
      "entropy": 5.613341426849365,
      "epoch": 0.6706866368410815,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960480072567557,
      "loss": 5.3829,
      "mean_token_accuracy": 0.17451770156621932,
      "num_tokens": 14952324.0,
      "step": 8620
    },
    {
      "entropy": 5.624604940414429,
      "epoch": 0.671075666212799,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004960428337949604,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1732550621032715,
      "num_tokens": 14961964.0,
      "step": 8625
    },
    {
      "entropy": 5.65596866607666,
      "epoch": 0.6714646955845166,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960376569791921,
      "loss": 5.3486,
      "mean_token_accuracy": 0.1749127060174942,
      "num_tokens": 14970267.0,
      "step": 8630
    },
    {
      "entropy": 5.4941013813018795,
      "epoch": 0.6718537249562342,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960324768095291,
      "loss": 5.2082,
      "mean_token_accuracy": 0.1824185848236084,
      "num_tokens": 14978546.0,
      "step": 8635
    },
    {
      "entropy": 5.591194486618042,
      "epoch": 0.6722427543279518,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960272932860502,
      "loss": 5.2791,
      "mean_token_accuracy": 0.17944162338972092,
      "num_tokens": 14986785.0,
      "step": 8640
    },
    {
      "entropy": 5.565518426895141,
      "epoch": 0.6726317836996694,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960221064088338,
      "loss": 5.2121,
      "mean_token_accuracy": 0.17982794344425201,
      "num_tokens": 14994873.0,
      "step": 8645
    },
    {
      "entropy": 5.528623771667481,
      "epoch": 0.6730208130713868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004960169161779588,
      "loss": 5.2442,
      "mean_token_accuracy": 0.18907042294740678,
      "num_tokens": 15003108.0,
      "step": 8650
    },
    {
      "entropy": 5.401618480682373,
      "epoch": 0.6734098424431044,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960117225935038,
      "loss": 5.1494,
      "mean_token_accuracy": 0.18058359175920485,
      "num_tokens": 15011802.0,
      "step": 8655
    },
    {
      "entropy": 5.530126285552979,
      "epoch": 0.673798871814822,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960065256555477,
      "loss": 5.3056,
      "mean_token_accuracy": 0.17976801544427873,
      "num_tokens": 15020785.0,
      "step": 8660
    },
    {
      "entropy": 5.646392774581909,
      "epoch": 0.6741879011865396,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960013253641695,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17093995064496995,
      "num_tokens": 15029218.0,
      "step": 8665
    },
    {
      "entropy": 5.60396032333374,
      "epoch": 0.6745769305582572,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959961217194477,
      "loss": 5.3492,
      "mean_token_accuracy": 0.16516607105731965,
      "num_tokens": 15037843.0,
      "step": 8670
    },
    {
      "entropy": 5.530879878997803,
      "epoch": 0.6749659599299748,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004959909147214615,
      "loss": 5.2725,
      "mean_token_accuracy": 0.18173045814037322,
      "num_tokens": 15047053.0,
      "step": 8675
    },
    {
      "entropy": 5.571678161621094,
      "epoch": 0.6753549893016922,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959857043702902,
      "loss": 5.3923,
      "mean_token_accuracy": 0.17267663776874542,
      "num_tokens": 15056159.0,
      "step": 8680
    },
    {
      "entropy": 5.560159969329834,
      "epoch": 0.6757440186734098,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959804906660124,
      "loss": 5.3358,
      "mean_token_accuracy": 0.16832655519247056,
      "num_tokens": 15064853.0,
      "step": 8685
    },
    {
      "entropy": 5.582832288742066,
      "epoch": 0.6761330480451274,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959752736087073,
      "loss": 5.3437,
      "mean_token_accuracy": 0.17521180510520934,
      "num_tokens": 15072671.0,
      "step": 8690
    },
    {
      "entropy": 5.525656604766846,
      "epoch": 0.676522077416845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959700531984543,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18609519004821778,
      "num_tokens": 15080787.0,
      "step": 8695
    },
    {
      "entropy": 5.537931632995606,
      "epoch": 0.6769111067885626,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959648294353324,
      "loss": 5.337,
      "mean_token_accuracy": 0.17343332022428512,
      "num_tokens": 15089906.0,
      "step": 8700
    },
    {
      "entropy": 5.553527593612671,
      "epoch": 0.6773001361602801,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959596023194208,
      "loss": 5.3828,
      "mean_token_accuracy": 0.17819689959287643,
      "num_tokens": 15099329.0,
      "step": 8705
    },
    {
      "entropy": 5.594941329956055,
      "epoch": 0.6776891655319977,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000495954371850799,
      "loss": 5.3673,
      "mean_token_accuracy": 0.17315385341644288,
      "num_tokens": 15107744.0,
      "step": 8710
    },
    {
      "entropy": 5.612883567810059,
      "epoch": 0.6780781949037152,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959491380295463,
      "loss": 5.383,
      "mean_token_accuracy": 0.1757282093167305,
      "num_tokens": 15115918.0,
      "step": 8715
    },
    {
      "entropy": 5.57036247253418,
      "epoch": 0.6784672242754328,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004959439008557422,
      "loss": 5.2432,
      "mean_token_accuracy": 0.18154032826423644,
      "num_tokens": 15124074.0,
      "step": 8720
    },
    {
      "entropy": 5.531103658676147,
      "epoch": 0.6788562536471504,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000495938660329466,
      "loss": 5.2554,
      "mean_token_accuracy": 0.18637112081050872,
      "num_tokens": 15132402.0,
      "step": 8725
    },
    {
      "entropy": 5.5278421401977536,
      "epoch": 0.6792452830188679,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004959334164507973,
      "loss": 5.381,
      "mean_token_accuracy": 0.17353248447179795,
      "num_tokens": 15141640.0,
      "step": 8730
    },
    {
      "entropy": 5.59519329071045,
      "epoch": 0.6796343123905855,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959281692198155,
      "loss": 5.261,
      "mean_token_accuracy": 0.18473790287971498,
      "num_tokens": 15149919.0,
      "step": 8735
    },
    {
      "entropy": 5.491541719436645,
      "epoch": 0.6800233417623031,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959229186366007,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19363221228122712,
      "num_tokens": 15157536.0,
      "step": 8740
    },
    {
      "entropy": 5.456545400619507,
      "epoch": 0.6804123711340206,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000495917664701232,
      "loss": 5.3419,
      "mean_token_accuracy": 0.17189399525523186,
      "num_tokens": 15165848.0,
      "step": 8745
    },
    {
      "entropy": 5.569458913803101,
      "epoch": 0.6808014005057382,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959124074137894,
      "loss": 5.324,
      "mean_token_accuracy": 0.17281854897737503,
      "num_tokens": 15174799.0,
      "step": 8750
    },
    {
      "entropy": 5.636410856246949,
      "epoch": 0.6811904298774557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959071467743526,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17031558603048325,
      "num_tokens": 15183695.0,
      "step": 8755
    },
    {
      "entropy": 5.664930534362793,
      "epoch": 0.6815794592491733,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004959018827830016,
      "loss": 5.4388,
      "mean_token_accuracy": 0.17168623954057693,
      "num_tokens": 15193620.0,
      "step": 8760
    },
    {
      "entropy": 5.611712694168091,
      "epoch": 0.6819684886208909,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495896615439816,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1811678811907768,
      "num_tokens": 15201535.0,
      "step": 8765
    },
    {
      "entropy": 5.654981660842895,
      "epoch": 0.6823575179926085,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958913447448759,
      "loss": 5.406,
      "mean_token_accuracy": 0.1682290881872177,
      "num_tokens": 15209678.0,
      "step": 8770
    },
    {
      "entropy": 5.486663866043091,
      "epoch": 0.682746547364326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958860706982613,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17819326519966125,
      "num_tokens": 15218579.0,
      "step": 8775
    },
    {
      "entropy": 5.437908411026001,
      "epoch": 0.6831355767360435,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000495880793300052,
      "loss": 5.2003,
      "mean_token_accuracy": 0.1872469514608383,
      "num_tokens": 15227225.0,
      "step": 8780
    },
    {
      "entropy": 5.575411701202393,
      "epoch": 0.6835246061077611,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004958755125503284,
      "loss": 5.3169,
      "mean_token_accuracy": 0.17595237642526626,
      "num_tokens": 15235900.0,
      "step": 8785
    },
    {
      "entropy": 5.615043497085571,
      "epoch": 0.6839136354794787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958702284491703,
      "loss": 5.2745,
      "mean_token_accuracy": 0.1787177413702011,
      "num_tokens": 15244538.0,
      "step": 8790
    },
    {
      "entropy": 5.476218605041504,
      "epoch": 0.6843026648511963,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004958649409966581,
      "loss": 5.2536,
      "mean_token_accuracy": 0.18225927352905275,
      "num_tokens": 15252902.0,
      "step": 8795
    },
    {
      "entropy": 5.539601230621338,
      "epoch": 0.6846916942229139,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495859650192872,
      "loss": 5.4068,
      "mean_token_accuracy": 0.16932478845119475,
      "num_tokens": 15261514.0,
      "step": 8800
    },
    {
      "entropy": 5.595240354537964,
      "epoch": 0.6850807235946313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004958543560378922,
      "loss": 5.3335,
      "mean_token_accuracy": 0.17627801597118378,
      "num_tokens": 15270140.0,
      "step": 8805
    },
    {
      "entropy": 5.552701330184936,
      "epoch": 0.6854697529663489,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004958490585317991,
      "loss": 5.2879,
      "mean_token_accuracy": 0.17693503946065903,
      "num_tokens": 15279551.0,
      "step": 8810
    },
    {
      "entropy": 5.494368410110473,
      "epoch": 0.6858587823380665,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495843757674673,
      "loss": 5.2693,
      "mean_token_accuracy": 0.1851952090859413,
      "num_tokens": 15288767.0,
      "step": 8815
    },
    {
      "entropy": 5.5630992412567135,
      "epoch": 0.6862478117097841,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958384534665945,
      "loss": 5.2907,
      "mean_token_accuracy": 0.1815930113196373,
      "num_tokens": 15297713.0,
      "step": 8820
    },
    {
      "entropy": 5.571929407119751,
      "epoch": 0.6866368410815017,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004958331459076438,
      "loss": 5.2981,
      "mean_token_accuracy": 0.17549618035554887,
      "num_tokens": 15306139.0,
      "step": 8825
    },
    {
      "entropy": 5.621170234680176,
      "epoch": 0.6870258704532192,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004958278349979018,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17428736686706542,
      "num_tokens": 15315015.0,
      "step": 8830
    },
    {
      "entropy": 5.564057683944702,
      "epoch": 0.6874148998249368,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958225207374488,
      "loss": 5.2324,
      "mean_token_accuracy": 0.1814328208565712,
      "num_tokens": 15323366.0,
      "step": 8835
    },
    {
      "entropy": 5.493619823455811,
      "epoch": 0.6878039291966543,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958172031263655,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1869863450527191,
      "num_tokens": 15331135.0,
      "step": 8840
    },
    {
      "entropy": 5.50973072052002,
      "epoch": 0.6881929585683719,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004958118821647326,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18131258487701415,
      "num_tokens": 15339561.0,
      "step": 8845
    },
    {
      "entropy": 5.563919305801392,
      "epoch": 0.6885819879400895,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004958065578526308,
      "loss": 5.146,
      "mean_token_accuracy": 0.18987538367509843,
      "num_tokens": 15348000.0,
      "step": 8850
    },
    {
      "entropy": 5.597847843170166,
      "epoch": 0.688971017311807,
      "grad_norm": 1.15625,
      "learning_rate": 0.000495801230190141,
      "loss": 5.3535,
      "mean_token_accuracy": 0.1803091585636139,
      "num_tokens": 15355916.0,
      "step": 8855
    },
    {
      "entropy": 5.417778778076172,
      "epoch": 0.6893600466835246,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004957958991773441,
      "loss": 5.1951,
      "mean_token_accuracy": 0.17953578978776932,
      "num_tokens": 15364561.0,
      "step": 8860
    },
    {
      "entropy": 5.5490752220153805,
      "epoch": 0.6897490760552422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957905648143206,
      "loss": 5.2359,
      "mean_token_accuracy": 0.17897392362356185,
      "num_tokens": 15372582.0,
      "step": 8865
    },
    {
      "entropy": 5.545386743545532,
      "epoch": 0.6901381054269597,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004957852271011519,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1711472824215889,
      "num_tokens": 15381684.0,
      "step": 8870
    },
    {
      "entropy": 5.6327790260314945,
      "epoch": 0.6905271347986773,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957798860379187,
      "loss": 5.3492,
      "mean_token_accuracy": 0.1704866975545883,
      "num_tokens": 15390858.0,
      "step": 8875
    },
    {
      "entropy": 5.549315643310547,
      "epoch": 0.6909161641703949,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004957745416247022,
      "loss": 5.175,
      "mean_token_accuracy": 0.18461504131555556,
      "num_tokens": 15399439.0,
      "step": 8880
    },
    {
      "entropy": 5.481848478317261,
      "epoch": 0.6913051935421124,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004957691938615833,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18324231952428818,
      "num_tokens": 15407784.0,
      "step": 8885
    },
    {
      "entropy": 5.619164657592774,
      "epoch": 0.69169422291383,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957638427486434,
      "loss": 5.3771,
      "mean_token_accuracy": 0.1737549901008606,
      "num_tokens": 15415985.0,
      "step": 8890
    },
    {
      "entropy": 5.524836540222168,
      "epoch": 0.6920832522855476,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004957584882859636,
      "loss": 5.2655,
      "mean_token_accuracy": 0.18107877522706986,
      "num_tokens": 15424433.0,
      "step": 8895
    },
    {
      "entropy": 5.529126119613648,
      "epoch": 0.6924722816572652,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004957531304736251,
      "loss": 5.3518,
      "mean_token_accuracy": 0.17498830556869507,
      "num_tokens": 15433041.0,
      "step": 8900
    },
    {
      "entropy": 5.493073415756226,
      "epoch": 0.6928613110289827,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004957477693117091,
      "loss": 5.1882,
      "mean_token_accuracy": 0.17712959349155427,
      "num_tokens": 15441159.0,
      "step": 8905
    },
    {
      "entropy": 5.6728880405426025,
      "epoch": 0.6932503404007002,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957424048002971,
      "loss": 5.503,
      "mean_token_accuracy": 0.17157228887081147,
      "num_tokens": 15450687.0,
      "step": 8910
    },
    {
      "entropy": 5.490257120132446,
      "epoch": 0.6936393697724178,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004957370369394706,
      "loss": 5.2117,
      "mean_token_accuracy": 0.1827506020665169,
      "num_tokens": 15460450.0,
      "step": 8915
    },
    {
      "entropy": 5.500311231613159,
      "epoch": 0.6940283991441354,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004957316657293107,
      "loss": 5.2392,
      "mean_token_accuracy": 0.18319796472787858,
      "num_tokens": 15468435.0,
      "step": 8920
    },
    {
      "entropy": 5.488282489776611,
      "epoch": 0.694417428515853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004957262911698992,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18449322432279586,
      "num_tokens": 15476232.0,
      "step": 8925
    },
    {
      "entropy": 5.606828832626343,
      "epoch": 0.6948064578875706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004957209132613175,
      "loss": 5.3058,
      "mean_token_accuracy": 0.181530599296093,
      "num_tokens": 15484569.0,
      "step": 8930
    },
    {
      "entropy": 5.541490125656128,
      "epoch": 0.695195487259288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957155320036473,
      "loss": 5.2239,
      "mean_token_accuracy": 0.1850089207291603,
      "num_tokens": 15492583.0,
      "step": 8935
    },
    {
      "entropy": 5.5534101009368895,
      "epoch": 0.6955845166310056,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957101473969702,
      "loss": 5.3658,
      "mean_token_accuracy": 0.17595504075288773,
      "num_tokens": 15501722.0,
      "step": 8940
    },
    {
      "entropy": 5.581349658966064,
      "epoch": 0.6959735460027232,
      "grad_norm": 1.109375,
      "learning_rate": 0.000495704759441368,
      "loss": 5.306,
      "mean_token_accuracy": 0.17931180596351623,
      "num_tokens": 15509682.0,
      "step": 8945
    },
    {
      "entropy": 5.620917654037475,
      "epoch": 0.6963625753744408,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956993681369221,
      "loss": 5.2986,
      "mean_token_accuracy": 0.18323905318975447,
      "num_tokens": 15518971.0,
      "step": 8950
    },
    {
      "entropy": 5.558902025222778,
      "epoch": 0.6967516047461584,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004956939734837148,
      "loss": 5.2752,
      "mean_token_accuracy": 0.17688121646642685,
      "num_tokens": 15527385.0,
      "step": 8955
    },
    {
      "entropy": 5.582807779312134,
      "epoch": 0.6971406341178759,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956885754818277,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17462659776210784,
      "num_tokens": 15535325.0,
      "step": 8960
    },
    {
      "entropy": 5.604357385635376,
      "epoch": 0.6975296634895934,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004956831741313427,
      "loss": 5.369,
      "mean_token_accuracy": 0.17730585783720015,
      "num_tokens": 15544098.0,
      "step": 8965
    },
    {
      "entropy": 5.5868736743927006,
      "epoch": 0.697918692861311,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004956777694323418,
      "loss": 5.2427,
      "mean_token_accuracy": 0.17640165984630585,
      "num_tokens": 15552547.0,
      "step": 8970
    },
    {
      "entropy": 5.513116931915283,
      "epoch": 0.6983077222330286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495672361384907,
      "loss": 5.2732,
      "mean_token_accuracy": 0.17443608194589616,
      "num_tokens": 15561044.0,
      "step": 8975
    },
    {
      "entropy": 5.55181188583374,
      "epoch": 0.6986967516047462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956669499891202,
      "loss": 5.3358,
      "mean_token_accuracy": 0.16730567812919617,
      "num_tokens": 15569379.0,
      "step": 8980
    },
    {
      "entropy": 5.604987478256225,
      "epoch": 0.6990857809764637,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956615352450639,
      "loss": 5.3403,
      "mean_token_accuracy": 0.1733030527830124,
      "num_tokens": 15578535.0,
      "step": 8985
    },
    {
      "entropy": 5.511005640029907,
      "epoch": 0.6994748103481813,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004956561171528198,
      "loss": 5.1889,
      "mean_token_accuracy": 0.18649003952741622,
      "num_tokens": 15586972.0,
      "step": 8990
    },
    {
      "entropy": 5.460616207122802,
      "epoch": 0.6998638397198989,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956506957124704,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1901437148451805,
      "num_tokens": 15595092.0,
      "step": 8995
    },
    {
      "entropy": 5.612773513793945,
      "epoch": 0.7002528690916164,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000495645270924098,
      "loss": 5.4611,
      "mean_token_accuracy": 0.17564321607351302,
      "num_tokens": 15604348.0,
      "step": 9000
    },
    {
      "epoch": 0.7002528690916164,
      "eval_entropy": 5.389798035222366,
      "eval_loss": 5.328912734985352,
      "eval_mean_token_accuracy": 0.18488086865384837,
      "eval_num_tokens": 15604348.0,
      "eval_runtime": 21.6858,
      "eval_samples_per_second": 1916.373,
      "eval_steps_per_second": 239.558,
      "step": 9000
    },
    {
      "entropy": 5.556111764907837,
      "epoch": 0.700641898463334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956398427877847,
      "loss": 5.3005,
      "mean_token_accuracy": 0.17604861706495284,
      "num_tokens": 15612597.0,
      "step": 9005
    },
    {
      "entropy": 5.540850639343262,
      "epoch": 0.7010309278350515,
      "grad_norm": 1.0,
      "learning_rate": 0.000495634411303613,
      "loss": 5.2751,
      "mean_token_accuracy": 0.1795080214738846,
      "num_tokens": 15621555.0,
      "step": 9010
    },
    {
      "entropy": 5.554760408401489,
      "epoch": 0.7014199572067691,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956289764716654,
      "loss": 5.2341,
      "mean_token_accuracy": 0.17902394235134125,
      "num_tokens": 15630518.0,
      "step": 9015
    },
    {
      "entropy": 5.5721453666687015,
      "epoch": 0.7018089865784867,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956235382920241,
      "loss": 5.41,
      "mean_token_accuracy": 0.16584509015083312,
      "num_tokens": 15639126.0,
      "step": 9020
    },
    {
      "entropy": 5.493175649642945,
      "epoch": 0.7021980159502043,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004956180967647717,
      "loss": 5.2275,
      "mean_token_accuracy": 0.1841141924262047,
      "num_tokens": 15647436.0,
      "step": 9025
    },
    {
      "entropy": 5.530618476867676,
      "epoch": 0.7025870453219218,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956126518899911,
      "loss": 5.2626,
      "mean_token_accuracy": 0.18042101860046386,
      "num_tokens": 15656439.0,
      "step": 9030
    },
    {
      "entropy": 5.552502059936524,
      "epoch": 0.7029760746936393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956072036677644,
      "loss": 5.2988,
      "mean_token_accuracy": 0.17644090354442596,
      "num_tokens": 15665119.0,
      "step": 9035
    },
    {
      "entropy": 5.566791677474976,
      "epoch": 0.7033651040653569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956017520981746,
      "loss": 5.3522,
      "mean_token_accuracy": 0.17432471066713334,
      "num_tokens": 15673430.0,
      "step": 9040
    },
    {
      "entropy": 5.643865871429443,
      "epoch": 0.7037541334370745,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004955962971813044,
      "loss": 5.4061,
      "mean_token_accuracy": 0.16947294473648072,
      "num_tokens": 15682739.0,
      "step": 9045
    },
    {
      "entropy": 5.558068943023682,
      "epoch": 0.7041431628087921,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004955908389172364,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1732873484492302,
      "num_tokens": 15691206.0,
      "step": 9050
    },
    {
      "entropy": 5.621509456634522,
      "epoch": 0.7045321921805097,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004955853773060536,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17490919977426528,
      "num_tokens": 15700238.0,
      "step": 9055
    },
    {
      "entropy": 5.600115966796875,
      "epoch": 0.7049212215522271,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004955799123478388,
      "loss": 5.2572,
      "mean_token_accuracy": 0.1805904373526573,
      "num_tokens": 15708654.0,
      "step": 9060
    },
    {
      "entropy": 5.606193733215332,
      "epoch": 0.7053102509239447,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955744440426748,
      "loss": 5.4156,
      "mean_token_accuracy": 0.17848946005105973,
      "num_tokens": 15717204.0,
      "step": 9065
    },
    {
      "entropy": 5.4877345085144045,
      "epoch": 0.7056992802956623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955689723906448,
      "loss": 5.1556,
      "mean_token_accuracy": 0.1923369437456131,
      "num_tokens": 15725016.0,
      "step": 9070
    },
    {
      "entropy": 5.649702692031861,
      "epoch": 0.7060883096673799,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004955634973918318,
      "loss": 5.5062,
      "mean_token_accuracy": 0.17271387353539466,
      "num_tokens": 15734714.0,
      "step": 9075
    },
    {
      "entropy": 5.596701908111572,
      "epoch": 0.7064773390390975,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955580190463186,
      "loss": 5.3283,
      "mean_token_accuracy": 0.17548872381448746,
      "num_tokens": 15742867.0,
      "step": 9080
    },
    {
      "entropy": 5.573616409301758,
      "epoch": 0.7068663684108151,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004955525373541886,
      "loss": 5.3359,
      "mean_token_accuracy": 0.17494852393865584,
      "num_tokens": 15752059.0,
      "step": 9085
    },
    {
      "entropy": 5.5120330333709715,
      "epoch": 0.7072553977825325,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004955470523155249,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17450406402349472,
      "num_tokens": 15761783.0,
      "step": 9090
    },
    {
      "entropy": 5.525177240371704,
      "epoch": 0.7076444271542501,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004955415639304107,
      "loss": 5.2011,
      "mean_token_accuracy": 0.1823938637971878,
      "num_tokens": 15770451.0,
      "step": 9095
    },
    {
      "entropy": 5.549926280975342,
      "epoch": 0.7080334565259677,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004955360721989292,
      "loss": 5.3088,
      "mean_token_accuracy": 0.18095069825649263,
      "num_tokens": 15779529.0,
      "step": 9100
    },
    {
      "entropy": 5.561652946472168,
      "epoch": 0.7084224858976853,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004955305771211641,
      "loss": 5.2761,
      "mean_token_accuracy": 0.1786726161837578,
      "num_tokens": 15788427.0,
      "step": 9105
    },
    {
      "entropy": 5.561768960952759,
      "epoch": 0.7088115152694029,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955250786971982,
      "loss": 5.3107,
      "mean_token_accuracy": 0.17775446325540542,
      "num_tokens": 15796530.0,
      "step": 9110
    },
    {
      "entropy": 5.49405255317688,
      "epoch": 0.7092005446411204,
      "grad_norm": 1.125,
      "learning_rate": 0.0004955195769271154,
      "loss": 5.2599,
      "mean_token_accuracy": 0.1851181998848915,
      "num_tokens": 15805058.0,
      "step": 9115
    },
    {
      "entropy": 5.65776515007019,
      "epoch": 0.709589574012838,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955140718109991,
      "loss": 5.3609,
      "mean_token_accuracy": 0.17976813912391662,
      "num_tokens": 15813729.0,
      "step": 9120
    },
    {
      "entropy": 5.577265214920044,
      "epoch": 0.7099786033845555,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004955085633489326,
      "loss": 5.2097,
      "mean_token_accuracy": 0.18789018839597701,
      "num_tokens": 15822305.0,
      "step": 9125
    },
    {
      "entropy": 5.553754806518555,
      "epoch": 0.7103676327562731,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004955030515409997,
      "loss": 5.3513,
      "mean_token_accuracy": 0.17451984584331512,
      "num_tokens": 15830842.0,
      "step": 9130
    },
    {
      "entropy": 5.469909429550171,
      "epoch": 0.7107566621279907,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004954975363872838,
      "loss": 5.1965,
      "mean_token_accuracy": 0.17950347810983658,
      "num_tokens": 15839790.0,
      "step": 9135
    },
    {
      "entropy": 5.5804030418396,
      "epoch": 0.7111456914997082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954920178878689,
      "loss": 5.3998,
      "mean_token_accuracy": 0.16854211539030076,
      "num_tokens": 15849307.0,
      "step": 9140
    },
    {
      "entropy": 5.647359418869018,
      "epoch": 0.7115347208714258,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004954864960428385,
      "loss": 5.3326,
      "mean_token_accuracy": 0.1800499364733696,
      "num_tokens": 15858688.0,
      "step": 9145
    },
    {
      "entropy": 5.576810455322265,
      "epoch": 0.7119237502431434,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954809708522765,
      "loss": 5.2099,
      "mean_token_accuracy": 0.17958054095506668,
      "num_tokens": 15867574.0,
      "step": 9150
    },
    {
      "entropy": 5.537421703338623,
      "epoch": 0.712312779614861,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954754423162667,
      "loss": 5.235,
      "mean_token_accuracy": 0.18324693739414216,
      "num_tokens": 15875871.0,
      "step": 9155
    },
    {
      "entropy": 5.529123449325562,
      "epoch": 0.7127018089865785,
      "grad_norm": 1.03125,
      "learning_rate": 0.000495469910434893,
      "loss": 5.303,
      "mean_token_accuracy": 0.17866899222135543,
      "num_tokens": 15884386.0,
      "step": 9160
    },
    {
      "entropy": 5.587394571304321,
      "epoch": 0.713090838358296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954643752082392,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17422557026147842,
      "num_tokens": 15893086.0,
      "step": 9165
    },
    {
      "entropy": 5.517653369903565,
      "epoch": 0.7134798677300136,
      "grad_norm": 1.0,
      "learning_rate": 0.0004954588366363896,
      "loss": 5.3217,
      "mean_token_accuracy": 0.17549112886190416,
      "num_tokens": 15902266.0,
      "step": 9170
    },
    {
      "entropy": 5.702410888671875,
      "epoch": 0.7138688971017312,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954532947194279,
      "loss": 5.448,
      "mean_token_accuracy": 0.16685094088315963,
      "num_tokens": 15911133.0,
      "step": 9175
    },
    {
      "entropy": 5.529084730148315,
      "epoch": 0.7142579264734488,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954477494574384,
      "loss": 5.2194,
      "mean_token_accuracy": 0.18191736340522766,
      "num_tokens": 15919455.0,
      "step": 9180
    },
    {
      "entropy": 5.49851655960083,
      "epoch": 0.7146469558451664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004954422008505052,
      "loss": 5.2822,
      "mean_token_accuracy": 0.18629485964775086,
      "num_tokens": 15927592.0,
      "step": 9185
    },
    {
      "entropy": 5.491498279571533,
      "epoch": 0.7150359852168838,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954366488987125,
      "loss": 5.221,
      "mean_token_accuracy": 0.1816892832517624,
      "num_tokens": 15935979.0,
      "step": 9190
    },
    {
      "entropy": 5.5549968719482425,
      "epoch": 0.7154250145886014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954310936021444,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17618247270584106,
      "num_tokens": 15945921.0,
      "step": 9195
    },
    {
      "entropy": 5.626093864440918,
      "epoch": 0.715814043960319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954255349608853,
      "loss": 5.3012,
      "mean_token_accuracy": 0.18052035868167876,
      "num_tokens": 15954875.0,
      "step": 9200
    },
    {
      "entropy": 5.56747088432312,
      "epoch": 0.7162030733320366,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954199729750198,
      "loss": 5.3087,
      "mean_token_accuracy": 0.17858943343162537,
      "num_tokens": 15963391.0,
      "step": 9205
    },
    {
      "entropy": 5.630508947372436,
      "epoch": 0.7165921027037542,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954144076446318,
      "loss": 5.427,
      "mean_token_accuracy": 0.16947771459817887,
      "num_tokens": 15972366.0,
      "step": 9210
    },
    {
      "entropy": 5.638298368453979,
      "epoch": 0.7169811320754716,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004954088389698061,
      "loss": 5.3947,
      "mean_token_accuracy": 0.177106873691082,
      "num_tokens": 15981196.0,
      "step": 9215
    },
    {
      "entropy": 5.48401198387146,
      "epoch": 0.7173701614471892,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495403266950627,
      "loss": 5.3291,
      "mean_token_accuracy": 0.17288484573364257,
      "num_tokens": 15990811.0,
      "step": 9220
    },
    {
      "entropy": 5.557605028152466,
      "epoch": 0.7177591908189068,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004953976915871792,
      "loss": 5.317,
      "mean_token_accuracy": 0.17596275806427003,
      "num_tokens": 15999594.0,
      "step": 9225
    },
    {
      "entropy": 5.687898397445679,
      "epoch": 0.7181482201906244,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004953921128795472,
      "loss": 5.4282,
      "mean_token_accuracy": 0.16423819810152054,
      "num_tokens": 16008179.0,
      "step": 9230
    },
    {
      "entropy": 5.582005643844605,
      "epoch": 0.718537249562342,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953865308278156,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17871031463146209,
      "num_tokens": 16016583.0,
      "step": 9235
    },
    {
      "entropy": 5.609802484512329,
      "epoch": 0.7189262789340595,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953809454320693,
      "loss": 5.356,
      "mean_token_accuracy": 0.17320025712251663,
      "num_tokens": 16025516.0,
      "step": 9240
    },
    {
      "entropy": 5.472203969955444,
      "epoch": 0.7193153083057771,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953753566923929,
      "loss": 5.2189,
      "mean_token_accuracy": 0.1882613107562065,
      "num_tokens": 16033536.0,
      "step": 9245
    },
    {
      "entropy": 5.519918298721313,
      "epoch": 0.7197043376774946,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953697646088712,
      "loss": 5.1493,
      "mean_token_accuracy": 0.1862490639090538,
      "num_tokens": 16041918.0,
      "step": 9250
    },
    {
      "entropy": 5.49515471458435,
      "epoch": 0.7200933670492122,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004953641691815891,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1890513315796852,
      "num_tokens": 16050806.0,
      "step": 9255
    },
    {
      "entropy": 5.549209785461426,
      "epoch": 0.7204823964209298,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004953585704106315,
      "loss": 5.3574,
      "mean_token_accuracy": 0.1688122883439064,
      "num_tokens": 16060140.0,
      "step": 9260
    },
    {
      "entropy": 5.590003824234008,
      "epoch": 0.7208714257926473,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004953529682960832,
      "loss": 5.2188,
      "mean_token_accuracy": 0.18012620657682418,
      "num_tokens": 16068182.0,
      "step": 9265
    },
    {
      "entropy": 5.591183280944824,
      "epoch": 0.7212604551643649,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004953473628380295,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1735826015472412,
      "num_tokens": 16076798.0,
      "step": 9270
    },
    {
      "entropy": 5.423460483551025,
      "epoch": 0.7216494845360825,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953417540365553,
      "loss": 5.1083,
      "mean_token_accuracy": 0.18438050895929337,
      "num_tokens": 16084739.0,
      "step": 9275
    },
    {
      "entropy": 5.559713363647461,
      "epoch": 0.7220385139078,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004953361418917455,
      "loss": 5.2661,
      "mean_token_accuracy": 0.18010660707950593,
      "num_tokens": 16093434.0,
      "step": 9280
    },
    {
      "entropy": 5.543549537658691,
      "epoch": 0.7224275432795176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004953305264036856,
      "loss": 5.3158,
      "mean_token_accuracy": 0.17837266623973846,
      "num_tokens": 16102163.0,
      "step": 9285
    },
    {
      "entropy": 5.489007902145386,
      "epoch": 0.7228165726512352,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953249075724607,
      "loss": 5.29,
      "mean_token_accuracy": 0.18584892600774766,
      "num_tokens": 16110761.0,
      "step": 9290
    },
    {
      "entropy": 5.545398187637329,
      "epoch": 0.7232056020229527,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004953192853981559,
      "loss": 5.325,
      "mean_token_accuracy": 0.1777096062898636,
      "num_tokens": 16119774.0,
      "step": 9295
    },
    {
      "entropy": 5.5385864734649655,
      "epoch": 0.7235946313946703,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004953136598808566,
      "loss": 5.196,
      "mean_token_accuracy": 0.18555736243724824,
      "num_tokens": 16128232.0,
      "step": 9300
    },
    {
      "entropy": 5.5455903053283695,
      "epoch": 0.7239836607663879,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004953080310206481,
      "loss": 5.2331,
      "mean_token_accuracy": 0.1827569469809532,
      "num_tokens": 16137370.0,
      "step": 9305
    },
    {
      "entropy": 5.507047891616821,
      "epoch": 0.7243726901381055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004953023988176162,
      "loss": 5.2536,
      "mean_token_accuracy": 0.1793384850025177,
      "num_tokens": 16145791.0,
      "step": 9310
    },
    {
      "entropy": 5.635753631591797,
      "epoch": 0.724761719509823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004952967632718458,
      "loss": 5.3666,
      "mean_token_accuracy": 0.17886020988225937,
      "num_tokens": 16153713.0,
      "step": 9315
    },
    {
      "entropy": 5.504679250717163,
      "epoch": 0.7251507488815405,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004952911243834227,
      "loss": 5.2166,
      "mean_token_accuracy": 0.17856226563453675,
      "num_tokens": 16162362.0,
      "step": 9320
    },
    {
      "entropy": 5.557938098907471,
      "epoch": 0.7255397782532581,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004952854821524324,
      "loss": 5.289,
      "mean_token_accuracy": 0.1809445485472679,
      "num_tokens": 16171094.0,
      "step": 9325
    },
    {
      "entropy": 5.59300479888916,
      "epoch": 0.7259288076249757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004952798365789606,
      "loss": 5.3487,
      "mean_token_accuracy": 0.17876584231853485,
      "num_tokens": 16180282.0,
      "step": 9330
    },
    {
      "entropy": 5.5803542137146,
      "epoch": 0.7263178369966933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004952741876630928,
      "loss": 5.2435,
      "mean_token_accuracy": 0.1806937649846077,
      "num_tokens": 16188497.0,
      "step": 9335
    },
    {
      "entropy": 5.607930040359497,
      "epoch": 0.7267068663684109,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952685354049148,
      "loss": 5.4144,
      "mean_token_accuracy": 0.16779697239398955,
      "num_tokens": 16197822.0,
      "step": 9340
    },
    {
      "entropy": 5.591944837570191,
      "epoch": 0.7270958957401283,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004952628798045124,
      "loss": 5.3089,
      "mean_token_accuracy": 0.17668944597244263,
      "num_tokens": 16207770.0,
      "step": 9345
    },
    {
      "entropy": 5.5045722961425785,
      "epoch": 0.7274849251118459,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952572208619714,
      "loss": 5.2394,
      "mean_token_accuracy": 0.1829061135649681,
      "num_tokens": 16216624.0,
      "step": 9350
    },
    {
      "entropy": 5.483622121810913,
      "epoch": 0.7278739544835635,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952515585773778,
      "loss": 5.1996,
      "mean_token_accuracy": 0.17681387364864348,
      "num_tokens": 16225585.0,
      "step": 9355
    },
    {
      "entropy": 5.52840633392334,
      "epoch": 0.7282629838552811,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952458929508171,
      "loss": 5.1684,
      "mean_token_accuracy": 0.18971401453018188,
      "num_tokens": 16233988.0,
      "step": 9360
    },
    {
      "entropy": 5.581288576126099,
      "epoch": 0.7286520132269987,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952402239823757,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17401268780231477,
      "num_tokens": 16242291.0,
      "step": 9365
    },
    {
      "entropy": 5.472625684738159,
      "epoch": 0.7290410425987162,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004952345516721393,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18057327717542648,
      "num_tokens": 16251102.0,
      "step": 9370
    },
    {
      "entropy": 5.479851818084716,
      "epoch": 0.7294300719704337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004952288760201942,
      "loss": 5.211,
      "mean_token_accuracy": 0.18755067735910416,
      "num_tokens": 16259136.0,
      "step": 9375
    },
    {
      "entropy": 5.559380006790161,
      "epoch": 0.7298191013421513,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004952231970266265,
      "loss": 5.3283,
      "mean_token_accuracy": 0.17745242565870284,
      "num_tokens": 16267958.0,
      "step": 9380
    },
    {
      "entropy": 5.667397785186767,
      "epoch": 0.7302081307138689,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952175146915223,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1704097494482994,
      "num_tokens": 16277616.0,
      "step": 9385
    },
    {
      "entropy": 5.65295000076294,
      "epoch": 0.7305971600855865,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004952118290149679,
      "loss": 5.3033,
      "mean_token_accuracy": 0.18273016065359116,
      "num_tokens": 16286454.0,
      "step": 9390
    },
    {
      "entropy": 5.566510534286499,
      "epoch": 0.730986189457304,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004952061399970492,
      "loss": 5.2741,
      "mean_token_accuracy": 0.17336290329694748,
      "num_tokens": 16295722.0,
      "step": 9395
    },
    {
      "entropy": 5.566493225097656,
      "epoch": 0.7313752188290216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495200447637853,
      "loss": 5.3381,
      "mean_token_accuracy": 0.17226790636777878,
      "num_tokens": 16304180.0,
      "step": 9400
    },
    {
      "entropy": 5.531685256958008,
      "epoch": 0.7317642482007392,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004951947519374655,
      "loss": 5.225,
      "mean_token_accuracy": 0.18427305519580842,
      "num_tokens": 16312766.0,
      "step": 9405
    },
    {
      "entropy": 5.545846557617187,
      "epoch": 0.7321532775724567,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951890528959731,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1751926675438881,
      "num_tokens": 16321174.0,
      "step": 9410
    },
    {
      "entropy": 5.498131275177002,
      "epoch": 0.7325423069441743,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951833505134623,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17664798051118852,
      "num_tokens": 16329870.0,
      "step": 9415
    },
    {
      "entropy": 5.5950346946716305,
      "epoch": 0.7329313363158918,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004951776447900196,
      "loss": 5.2883,
      "mean_token_accuracy": 0.17546151727437972,
      "num_tokens": 16338998.0,
      "step": 9420
    },
    {
      "entropy": 5.59651255607605,
      "epoch": 0.7333203656876094,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951719357257317,
      "loss": 5.2742,
      "mean_token_accuracy": 0.17708998173475266,
      "num_tokens": 16347313.0,
      "step": 9425
    },
    {
      "entropy": 5.56943187713623,
      "epoch": 0.733709395059327,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000495166223320685,
      "loss": 5.3988,
      "mean_token_accuracy": 0.17598551213741304,
      "num_tokens": 16356565.0,
      "step": 9430
    },
    {
      "entropy": 5.5626061916351315,
      "epoch": 0.7340984244310446,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951605075749662,
      "loss": 5.3924,
      "mean_token_accuracy": 0.1750992491841316,
      "num_tokens": 16365747.0,
      "step": 9435
    },
    {
      "entropy": 5.606553077697754,
      "epoch": 0.7344874538027621,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951547884886623,
      "loss": 5.3413,
      "mean_token_accuracy": 0.1672863632440567,
      "num_tokens": 16374650.0,
      "step": 9440
    },
    {
      "entropy": 5.567455911636353,
      "epoch": 0.7348764831744796,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004951490660618598,
      "loss": 5.2837,
      "mean_token_accuracy": 0.17836343348026276,
      "num_tokens": 16384119.0,
      "step": 9445
    },
    {
      "entropy": 5.544294929504394,
      "epoch": 0.7352655125461972,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951433402946457,
      "loss": 5.2417,
      "mean_token_accuracy": 0.18445164561271668,
      "num_tokens": 16392606.0,
      "step": 9450
    },
    {
      "entropy": 5.481019306182861,
      "epoch": 0.7356545419179148,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951376111871068,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18546178191900253,
      "num_tokens": 16401006.0,
      "step": 9455
    },
    {
      "entropy": 5.523701953887939,
      "epoch": 0.7360435712896324,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004951318787393299,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18083978742361068,
      "num_tokens": 16409401.0,
      "step": 9460
    },
    {
      "entropy": 5.602548837661743,
      "epoch": 0.73643260066135,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951261429514023,
      "loss": 5.3174,
      "mean_token_accuracy": 0.17583308815956117,
      "num_tokens": 16418771.0,
      "step": 9465
    },
    {
      "entropy": 5.5785486698150635,
      "epoch": 0.7368216300330674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951204038234106,
      "loss": 5.3619,
      "mean_token_accuracy": 0.1764096811413765,
      "num_tokens": 16427624.0,
      "step": 9470
    },
    {
      "entropy": 5.541777753829956,
      "epoch": 0.737210659404785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004951146613554424,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1854480504989624,
      "num_tokens": 16435669.0,
      "step": 9475
    },
    {
      "entropy": 5.582617092132568,
      "epoch": 0.7375996887765026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004951089155475843,
      "loss": 5.2911,
      "mean_token_accuracy": 0.18146839141845703,
      "num_tokens": 16444283.0,
      "step": 9480
    },
    {
      "entropy": 5.4931660175323485,
      "epoch": 0.7379887181482202,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495103166399924,
      "loss": 5.2616,
      "mean_token_accuracy": 0.17441291362047195,
      "num_tokens": 16452741.0,
      "step": 9485
    },
    {
      "entropy": 5.522454881668091,
      "epoch": 0.7383777475199378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004950974139125484,
      "loss": 5.3088,
      "mean_token_accuracy": 0.1850725769996643,
      "num_tokens": 16461386.0,
      "step": 9490
    },
    {
      "entropy": 5.627692127227784,
      "epoch": 0.7387667768916554,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950916580855448,
      "loss": 5.3251,
      "mean_token_accuracy": 0.17463397830724717,
      "num_tokens": 16469901.0,
      "step": 9495
    },
    {
      "entropy": 5.560376739501953,
      "epoch": 0.7391558062633728,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004950858989190007,
      "loss": 5.2499,
      "mean_token_accuracy": 0.1811629593372345,
      "num_tokens": 16478076.0,
      "step": 9500
    },
    {
      "entropy": 5.480900526046753,
      "epoch": 0.7395448356350904,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950801364130032,
      "loss": 5.2604,
      "mean_token_accuracy": 0.18097128719091415,
      "num_tokens": 16486568.0,
      "step": 9505
    },
    {
      "entropy": 5.5960197925567625,
      "epoch": 0.739933865006808,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950743705676401,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17725475132465363,
      "num_tokens": 16495299.0,
      "step": 9510
    },
    {
      "entropy": 5.539008760452271,
      "epoch": 0.7403228943785256,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004950686013829987,
      "loss": 5.2531,
      "mean_token_accuracy": 0.17483892142772675,
      "num_tokens": 16504576.0,
      "step": 9515
    },
    {
      "entropy": 5.574979066848755,
      "epoch": 0.7407119237502432,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004950628288591665,
      "loss": 5.3242,
      "mean_token_accuracy": 0.17738772332668304,
      "num_tokens": 16512715.0,
      "step": 9520
    },
    {
      "entropy": 5.613685178756714,
      "epoch": 0.7411009531219607,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950570529962311,
      "loss": 5.2746,
      "mean_token_accuracy": 0.1810723900794983,
      "num_tokens": 16521375.0,
      "step": 9525
    },
    {
      "entropy": 5.55305233001709,
      "epoch": 0.7414899824936783,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004950512737942803,
      "loss": 5.2493,
      "mean_token_accuracy": 0.18598097711801528,
      "num_tokens": 16529772.0,
      "step": 9530
    },
    {
      "entropy": 5.611551809310913,
      "epoch": 0.7418790118653958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950454912534015,
      "loss": 5.2967,
      "mean_token_accuracy": 0.18408222794532775,
      "num_tokens": 16538365.0,
      "step": 9535
    },
    {
      "entropy": 5.6260138034820555,
      "epoch": 0.7422680412371134,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950397053736827,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16625734269618989,
      "num_tokens": 16547668.0,
      "step": 9540
    },
    {
      "entropy": 5.631437683105469,
      "epoch": 0.742657070608831,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004950339161552117,
      "loss": 5.3923,
      "mean_token_accuracy": 0.16748718023300171,
      "num_tokens": 16557060.0,
      "step": 9545
    },
    {
      "entropy": 5.593369626998902,
      "epoch": 0.7430460999805485,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004950281235980761,
      "loss": 5.3649,
      "mean_token_accuracy": 0.18361901491880417,
      "num_tokens": 16565814.0,
      "step": 9550
    },
    {
      "entropy": 5.5311542510986325,
      "epoch": 0.7434351293522661,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495022327702364,
      "loss": 5.1699,
      "mean_token_accuracy": 0.18616349399089813,
      "num_tokens": 16574536.0,
      "step": 9555
    },
    {
      "entropy": 5.522435569763184,
      "epoch": 0.7438241587239837,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950165284681631,
      "loss": 5.1448,
      "mean_token_accuracy": 0.19001736789941787,
      "num_tokens": 16583349.0,
      "step": 9560
    },
    {
      "entropy": 5.485368776321411,
      "epoch": 0.7442131880957013,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004950107258955618,
      "loss": 5.1509,
      "mean_token_accuracy": 0.18268264830112457,
      "num_tokens": 16591374.0,
      "step": 9565
    },
    {
      "entropy": 5.503294038772583,
      "epoch": 0.7446022174674188,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950049199846479,
      "loss": 5.2609,
      "mean_token_accuracy": 0.17910871505737305,
      "num_tokens": 16599684.0,
      "step": 9570
    },
    {
      "entropy": 5.556149101257324,
      "epoch": 0.7449912468391363,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004949991107355095,
      "loss": 5.2553,
      "mean_token_accuracy": 0.17982286363840103,
      "num_tokens": 16607618.0,
      "step": 9575
    },
    {
      "entropy": 5.535337591171265,
      "epoch": 0.7453802762108539,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949932981482347,
      "loss": 5.3675,
      "mean_token_accuracy": 0.17090606391429902,
      "num_tokens": 16616259.0,
      "step": 9580
    },
    {
      "entropy": 5.541644048690796,
      "epoch": 0.7457693055825715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004949874822229118,
      "loss": 5.225,
      "mean_token_accuracy": 0.1834418684244156,
      "num_tokens": 16624972.0,
      "step": 9585
    },
    {
      "entropy": 5.61938967704773,
      "epoch": 0.7461583349542891,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949816629596288,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17839625775814055,
      "num_tokens": 16633517.0,
      "step": 9590
    },
    {
      "entropy": 5.598763275146484,
      "epoch": 0.7465473643260067,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004949758403584746,
      "loss": 5.3331,
      "mean_token_accuracy": 0.1768048346042633,
      "num_tokens": 16642695.0,
      "step": 9595
    },
    {
      "entropy": 5.459473657608032,
      "epoch": 0.7469363936977241,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949700144195368,
      "loss": 5.2168,
      "mean_token_accuracy": 0.1797614425420761,
      "num_tokens": 16651251.0,
      "step": 9600
    },
    {
      "entropy": 5.521040964126587,
      "epoch": 0.7473254230694417,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949641851429043,
      "loss": 5.2773,
      "mean_token_accuracy": 0.17668697834014893,
      "num_tokens": 16660570.0,
      "step": 9605
    },
    {
      "entropy": 5.5326752185821535,
      "epoch": 0.7477144524411593,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949583525286654,
      "loss": 5.312,
      "mean_token_accuracy": 0.17187793254852296,
      "num_tokens": 16668678.0,
      "step": 9610
    },
    {
      "entropy": 5.43628339767456,
      "epoch": 0.7481034818128769,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004949525165769085,
      "loss": 5.1444,
      "mean_token_accuracy": 0.18645231425762177,
      "num_tokens": 16677207.0,
      "step": 9615
    },
    {
      "entropy": 5.582274103164673,
      "epoch": 0.7484925111845945,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004949466772877224,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1755437359213829,
      "num_tokens": 16686015.0,
      "step": 9620
    },
    {
      "entropy": 5.54477105140686,
      "epoch": 0.748881540556312,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004949408346611955,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18544931560754777,
      "num_tokens": 16694989.0,
      "step": 9625
    },
    {
      "entropy": 5.584083795547485,
      "epoch": 0.7492705699280295,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004949349886974165,
      "loss": 5.3611,
      "mean_token_accuracy": 0.17635638415813445,
      "num_tokens": 16703888.0,
      "step": 9630
    },
    {
      "entropy": 5.55716118812561,
      "epoch": 0.7496595992997471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004949291393964741,
      "loss": 5.2835,
      "mean_token_accuracy": 0.17723703235387803,
      "num_tokens": 16711880.0,
      "step": 9635
    },
    {
      "entropy": 5.583466053009033,
      "epoch": 0.7500486286714647,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494923286758457,
      "loss": 5.2772,
      "mean_token_accuracy": 0.18172440379858018,
      "num_tokens": 16720858.0,
      "step": 9640
    },
    {
      "entropy": 5.550045967102051,
      "epoch": 0.7504376580431823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004949174307834541,
      "loss": 5.3488,
      "mean_token_accuracy": 0.17593863606452942,
      "num_tokens": 16729223.0,
      "step": 9645
    },
    {
      "entropy": 5.578950548171997,
      "epoch": 0.7508266874148998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949115714715543,
      "loss": 5.316,
      "mean_token_accuracy": 0.17637102603912352,
      "num_tokens": 16737854.0,
      "step": 9650
    },
    {
      "entropy": 5.5745062828063965,
      "epoch": 0.7512157167866174,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949057088228465,
      "loss": 5.3249,
      "mean_token_accuracy": 0.17119787484407425,
      "num_tokens": 16747081.0,
      "step": 9655
    },
    {
      "entropy": 5.498596286773681,
      "epoch": 0.751604746158335,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948998428374194,
      "loss": 5.1653,
      "mean_token_accuracy": 0.18802380561828613,
      "num_tokens": 16756410.0,
      "step": 9660
    },
    {
      "entropy": 5.494810676574707,
      "epoch": 0.7519937755300525,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948939735153622,
      "loss": 5.1923,
      "mean_token_accuracy": 0.18529070615768434,
      "num_tokens": 16764774.0,
      "step": 9665
    },
    {
      "entropy": 5.528837108612061,
      "epoch": 0.7523828049017701,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948881008567641,
      "loss": 5.3482,
      "mean_token_accuracy": 0.17671961337327957,
      "num_tokens": 16773484.0,
      "step": 9670
    },
    {
      "entropy": 5.576663970947266,
      "epoch": 0.7527718342734876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004948822248617139,
      "loss": 5.1586,
      "mean_token_accuracy": 0.191119584441185,
      "num_tokens": 16781167.0,
      "step": 9675
    },
    {
      "entropy": 5.464426851272583,
      "epoch": 0.7531608636452052,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000494876345530301,
      "loss": 5.212,
      "mean_token_accuracy": 0.18175776600837706,
      "num_tokens": 16790838.0,
      "step": 9680
    },
    {
      "entropy": 5.524459981918335,
      "epoch": 0.7535498930169228,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948704628626145,
      "loss": 5.2641,
      "mean_token_accuracy": 0.1849897548556328,
      "num_tokens": 16799179.0,
      "step": 9685
    },
    {
      "entropy": 5.606106233596802,
      "epoch": 0.7539389223886404,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948645768587437,
      "loss": 5.1909,
      "mean_token_accuracy": 0.17714202255010605,
      "num_tokens": 16807516.0,
      "step": 9690
    },
    {
      "entropy": 5.601245403289795,
      "epoch": 0.7543279517603579,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004948586875187778,
      "loss": 5.4131,
      "mean_token_accuracy": 0.17605143338441848,
      "num_tokens": 16816601.0,
      "step": 9695
    },
    {
      "entropy": 5.533262538909912,
      "epoch": 0.7547169811320755,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004948527948428064,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17988608181476592,
      "num_tokens": 16825498.0,
      "step": 9700
    },
    {
      "entropy": 5.593303298950195,
      "epoch": 0.755106010503793,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948468988309187,
      "loss": 5.3737,
      "mean_token_accuracy": 0.17876372784376143,
      "num_tokens": 16834833.0,
      "step": 9705
    },
    {
      "entropy": 5.560336494445801,
      "epoch": 0.7554950398755106,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004948409994832043,
      "loss": 5.2511,
      "mean_token_accuracy": 0.18087394088506697,
      "num_tokens": 16843476.0,
      "step": 9710
    },
    {
      "entropy": 5.517852830886841,
      "epoch": 0.7558840692472282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004948350967997526,
      "loss": 5.1705,
      "mean_token_accuracy": 0.18963382989168168,
      "num_tokens": 16851859.0,
      "step": 9715
    },
    {
      "entropy": 5.580209112167358,
      "epoch": 0.7562730986189458,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004948291907806532,
      "loss": 5.2378,
      "mean_token_accuracy": 0.18185634762048722,
      "num_tokens": 16860642.0,
      "step": 9720
    },
    {
      "entropy": 5.591513061523438,
      "epoch": 0.7566621279906633,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948232814259957,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1745381847023964,
      "num_tokens": 16869232.0,
      "step": 9725
    },
    {
      "entropy": 5.492669534683228,
      "epoch": 0.7570511573623808,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004948173687358699,
      "loss": 5.1244,
      "mean_token_accuracy": 0.1836826354265213,
      "num_tokens": 16877880.0,
      "step": 9730
    },
    {
      "entropy": 5.456797504425049,
      "epoch": 0.7574401867340984,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004948114527103652,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18095895498991013,
      "num_tokens": 16887060.0,
      "step": 9735
    },
    {
      "entropy": 5.574520063400269,
      "epoch": 0.757829216105816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948055333495717,
      "loss": 5.2626,
      "mean_token_accuracy": 0.17601558417081833,
      "num_tokens": 16895980.0,
      "step": 9740
    },
    {
      "entropy": 5.662924909591675,
      "epoch": 0.7582182454775336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004947996106535791,
      "loss": 5.4047,
      "mean_token_accuracy": 0.17123472839593887,
      "num_tokens": 16904334.0,
      "step": 9745
    },
    {
      "entropy": 5.501442003250122,
      "epoch": 0.7586072748492512,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004947936846224773,
      "loss": 5.2864,
      "mean_token_accuracy": 0.18169119507074355,
      "num_tokens": 16913436.0,
      "step": 9750
    },
    {
      "entropy": 5.503163862228393,
      "epoch": 0.7589963042209686,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947877552563562,
      "loss": 5.2727,
      "mean_token_accuracy": 0.17634390145540238,
      "num_tokens": 16923214.0,
      "step": 9755
    },
    {
      "entropy": 5.579230499267578,
      "epoch": 0.7593853335926862,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004947818225553056,
      "loss": 5.2044,
      "mean_token_accuracy": 0.18720707297325134,
      "num_tokens": 16932523.0,
      "step": 9760
    },
    {
      "entropy": 5.601711845397949,
      "epoch": 0.7597743629644038,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947758865194156,
      "loss": 5.3313,
      "mean_token_accuracy": 0.17638456970453262,
      "num_tokens": 16940649.0,
      "step": 9765
    },
    {
      "entropy": 5.550697326660156,
      "epoch": 0.7601633923361214,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004947699471487766,
      "loss": 5.2834,
      "mean_token_accuracy": 0.18467652052640915,
      "num_tokens": 16949373.0,
      "step": 9770
    },
    {
      "entropy": 5.604041671752929,
      "epoch": 0.760552421707839,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004947640044434782,
      "loss": 5.2943,
      "mean_token_accuracy": 0.18377356231212616,
      "num_tokens": 16957787.0,
      "step": 9775
    },
    {
      "entropy": 5.525785446166992,
      "epoch": 0.7609414510795565,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004947580584036109,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18299616128206253,
      "num_tokens": 16967589.0,
      "step": 9780
    },
    {
      "entropy": 5.551060009002685,
      "epoch": 0.761330480451274,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004947521090292649,
      "loss": 5.2228,
      "mean_token_accuracy": 0.18813629150390626,
      "num_tokens": 16976719.0,
      "step": 9785
    },
    {
      "entropy": 5.5405426025390625,
      "epoch": 0.7617195098229916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004947461563205304,
      "loss": 5.3578,
      "mean_token_accuracy": 0.18054259270429612,
      "num_tokens": 16984805.0,
      "step": 9790
    },
    {
      "entropy": 5.530938386917114,
      "epoch": 0.7621085391947092,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947402002774977,
      "loss": 5.2936,
      "mean_token_accuracy": 0.175130957365036,
      "num_tokens": 16993298.0,
      "step": 9795
    },
    {
      "entropy": 5.506849193572998,
      "epoch": 0.7624975685664268,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947342409002572,
      "loss": 5.1691,
      "mean_token_accuracy": 0.19266698211431504,
      "num_tokens": 17002170.0,
      "step": 9800
    },
    {
      "entropy": 5.521035671234131,
      "epoch": 0.7628865979381443,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947282781888994,
      "loss": 5.2667,
      "mean_token_accuracy": 0.19202028065919877,
      "num_tokens": 17011218.0,
      "step": 9805
    },
    {
      "entropy": 5.520613718032837,
      "epoch": 0.7632756273098619,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004947223121435147,
      "loss": 5.1636,
      "mean_token_accuracy": 0.1823309525847435,
      "num_tokens": 17019628.0,
      "step": 9810
    },
    {
      "entropy": 5.545638942718506,
      "epoch": 0.7636646566815795,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004947163427641936,
      "loss": 5.3798,
      "mean_token_accuracy": 0.17455653548240663,
      "num_tokens": 17028342.0,
      "step": 9815
    },
    {
      "entropy": 5.6043336391448975,
      "epoch": 0.764053686053297,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947103700510268,
      "loss": 5.3258,
      "mean_token_accuracy": 0.176914082467556,
      "num_tokens": 17037844.0,
      "step": 9820
    },
    {
      "entropy": 5.538714170455933,
      "epoch": 0.7644427154250146,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947043940041047,
      "loss": 5.1678,
      "mean_token_accuracy": 0.18721466660499572,
      "num_tokens": 17046430.0,
      "step": 9825
    },
    {
      "entropy": 5.522770977020263,
      "epoch": 0.7648317447967321,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004946984146235183,
      "loss": 5.2463,
      "mean_token_accuracy": 0.17984112799167634,
      "num_tokens": 17055757.0,
      "step": 9830
    },
    {
      "entropy": 5.562282133102417,
      "epoch": 0.7652207741684497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004946924319093579,
      "loss": 5.3668,
      "mean_token_accuracy": 0.1770143076777458,
      "num_tokens": 17063944.0,
      "step": 9835
    },
    {
      "entropy": 5.570206069946289,
      "epoch": 0.7656098035401673,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946864458617148,
      "loss": 5.2588,
      "mean_token_accuracy": 0.17300845831632614,
      "num_tokens": 17072895.0,
      "step": 9840
    },
    {
      "entropy": 5.521096086502075,
      "epoch": 0.7659988329118849,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946804564806793,
      "loss": 5.2973,
      "mean_token_accuracy": 0.1775191456079483,
      "num_tokens": 17081828.0,
      "step": 9845
    },
    {
      "entropy": 5.536095046997071,
      "epoch": 0.7663878622836025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946744637663427,
      "loss": 5.2203,
      "mean_token_accuracy": 0.18323799669742585,
      "num_tokens": 17090386.0,
      "step": 9850
    },
    {
      "entropy": 5.663650846481323,
      "epoch": 0.7667768916553199,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946684677187956,
      "loss": 5.4161,
      "mean_token_accuracy": 0.1748966947197914,
      "num_tokens": 17099429.0,
      "step": 9855
    },
    {
      "entropy": 5.554952335357666,
      "epoch": 0.7671659210270375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946624683381292,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17905012965202333,
      "num_tokens": 17107868.0,
      "step": 9860
    },
    {
      "entropy": 5.5565272808074955,
      "epoch": 0.7675549503987551,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004946564656244345,
      "loss": 5.3356,
      "mean_token_accuracy": 0.17248685657978058,
      "num_tokens": 17116936.0,
      "step": 9865
    },
    {
      "entropy": 5.590495014190674,
      "epoch": 0.7679439797704727,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004946504595778026,
      "loss": 5.264,
      "mean_token_accuracy": 0.17216774672269822,
      "num_tokens": 17125475.0,
      "step": 9870
    },
    {
      "entropy": 5.595300626754761,
      "epoch": 0.7683330091421903,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004946444501983246,
      "loss": 5.2723,
      "mean_token_accuracy": 0.18637385070323945,
      "num_tokens": 17134041.0,
      "step": 9875
    },
    {
      "entropy": 5.534225416183472,
      "epoch": 0.7687220385139077,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946384374860916,
      "loss": 5.2857,
      "mean_token_accuracy": 0.1820813775062561,
      "num_tokens": 17142088.0,
      "step": 9880
    },
    {
      "entropy": 5.5372497081756595,
      "epoch": 0.7691110678856253,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946324214411949,
      "loss": 5.1765,
      "mean_token_accuracy": 0.1841478690505028,
      "num_tokens": 17150185.0,
      "step": 9885
    },
    {
      "entropy": 5.470294761657715,
      "epoch": 0.7695000972573429,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004946264020637259,
      "loss": 5.0628,
      "mean_token_accuracy": 0.1939447656273842,
      "num_tokens": 17158437.0,
      "step": 9890
    },
    {
      "entropy": 5.440548849105835,
      "epoch": 0.7698891266290605,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946203793537757,
      "loss": 5.244,
      "mean_token_accuracy": 0.1837555557489395,
      "num_tokens": 17167811.0,
      "step": 9895
    },
    {
      "entropy": 5.562580537796021,
      "epoch": 0.7702781560007781,
      "grad_norm": 1.0,
      "learning_rate": 0.0004946143533114358,
      "loss": 5.3101,
      "mean_token_accuracy": 0.17614355236291884,
      "num_tokens": 17176982.0,
      "step": 9900
    },
    {
      "entropy": 5.608755016326905,
      "epoch": 0.7706671853724957,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004946083239367976,
      "loss": 5.2377,
      "mean_token_accuracy": 0.17974676787853242,
      "num_tokens": 17185633.0,
      "step": 9905
    },
    {
      "entropy": 5.4947271823883055,
      "epoch": 0.7710562147442132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004946022912299527,
      "loss": 5.285,
      "mean_token_accuracy": 0.17526287883520125,
      "num_tokens": 17194996.0,
      "step": 9910
    },
    {
      "entropy": 5.6518114566802975,
      "epoch": 0.7714452441159307,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945962551909926,
      "loss": 5.3927,
      "mean_token_accuracy": 0.17043281644582747,
      "num_tokens": 17204159.0,
      "step": 9915
    },
    {
      "entropy": 5.51574559211731,
      "epoch": 0.7718342734876483,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945902158200089,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1770494043827057,
      "num_tokens": 17212930.0,
      "step": 9920
    },
    {
      "entropy": 5.624597072601318,
      "epoch": 0.7722233028593659,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004945841731170931,
      "loss": 5.3393,
      "mean_token_accuracy": 0.17416706383228303,
      "num_tokens": 17221884.0,
      "step": 9925
    },
    {
      "entropy": 5.518691301345825,
      "epoch": 0.7726123322310835,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945781270823369,
      "loss": 5.2288,
      "mean_token_accuracy": 0.18571600914001465,
      "num_tokens": 17230525.0,
      "step": 9930
    },
    {
      "entropy": 5.439535808563233,
      "epoch": 0.773001361602801,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945720777158323,
      "loss": 5.1666,
      "mean_token_accuracy": 0.18698474764823914,
      "num_tokens": 17239240.0,
      "step": 9935
    },
    {
      "entropy": 5.619084358215332,
      "epoch": 0.7733903909745186,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004945660250176708,
      "loss": 5.2884,
      "mean_token_accuracy": 0.1752843976020813,
      "num_tokens": 17248174.0,
      "step": 9940
    },
    {
      "entropy": 5.574867486953735,
      "epoch": 0.7737794203462361,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004945599689879443,
      "loss": 5.2803,
      "mean_token_accuracy": 0.17735401540994644,
      "num_tokens": 17256935.0,
      "step": 9945
    },
    {
      "entropy": 5.5688135623931885,
      "epoch": 0.7741684497179537,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945539096267448,
      "loss": 5.2913,
      "mean_token_accuracy": 0.18053762316703797,
      "num_tokens": 17265831.0,
      "step": 9950
    },
    {
      "entropy": 5.469559907913208,
      "epoch": 0.7745574790896713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004945478469341642,
      "loss": 5.1775,
      "mean_token_accuracy": 0.188815376162529,
      "num_tokens": 17273851.0,
      "step": 9955
    },
    {
      "entropy": 5.547821998596191,
      "epoch": 0.7749465084613888,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945417809102944,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17088476270437242,
      "num_tokens": 17283547.0,
      "step": 9960
    },
    {
      "entropy": 5.490760421752929,
      "epoch": 0.7753355378331064,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945357115552275,
      "loss": 5.1435,
      "mean_token_accuracy": 0.18845568150281905,
      "num_tokens": 17291958.0,
      "step": 9965
    },
    {
      "entropy": 5.469471740722656,
      "epoch": 0.775724567204824,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004945296388690557,
      "loss": 5.2974,
      "mean_token_accuracy": 0.1850552275776863,
      "num_tokens": 17301374.0,
      "step": 9970
    },
    {
      "entropy": 5.597596454620361,
      "epoch": 0.7761135965765416,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000494523562851871,
      "loss": 5.4476,
      "mean_token_accuracy": 0.17390787452459336,
      "num_tokens": 17310649.0,
      "step": 9975
    },
    {
      "entropy": 5.664722537994384,
      "epoch": 0.7765026259482591,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945174835037655,
      "loss": 5.3408,
      "mean_token_accuracy": 0.1781105414032936,
      "num_tokens": 17319156.0,
      "step": 9980
    },
    {
      "entropy": 5.5528970718383786,
      "epoch": 0.7768916553199766,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004945114008248319,
      "loss": 5.2616,
      "mean_token_accuracy": 0.18411820977926255,
      "num_tokens": 17327906.0,
      "step": 9985
    },
    {
      "entropy": 5.530083465576172,
      "epoch": 0.7772806846916942,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004945053148151619,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1852427065372467,
      "num_tokens": 17337042.0,
      "step": 9990
    },
    {
      "entropy": 5.544866991043091,
      "epoch": 0.7776697140634118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944992254748482,
      "loss": 5.2979,
      "mean_token_accuracy": 0.17376251220703126,
      "num_tokens": 17346208.0,
      "step": 9995
    },
    {
      "entropy": 5.609505224227905,
      "epoch": 0.7780587434351294,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944931328039832,
      "loss": 5.3232,
      "mean_token_accuracy": 0.18415360450744628,
      "num_tokens": 17354965.0,
      "step": 10000
    },
    {
      "entropy": 5.57321515083313,
      "epoch": 0.778447772806847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944870368026592,
      "loss": 5.284,
      "mean_token_accuracy": 0.17819664627313614,
      "num_tokens": 17364129.0,
      "step": 10005
    },
    {
      "entropy": 5.573775291442871,
      "epoch": 0.7788368021785644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944809374709689,
      "loss": 5.2878,
      "mean_token_accuracy": 0.17610488533973695,
      "num_tokens": 17372928.0,
      "step": 10010
    },
    {
      "entropy": 5.612327718734742,
      "epoch": 0.779225831550282,
      "grad_norm": 1.125,
      "learning_rate": 0.0004944748348090045,
      "loss": 5.3089,
      "mean_token_accuracy": 0.17743565291166305,
      "num_tokens": 17381076.0,
      "step": 10015
    },
    {
      "entropy": 5.590665245056153,
      "epoch": 0.7796148609219996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004944687288168589,
      "loss": 5.3056,
      "mean_token_accuracy": 0.1801182821393013,
      "num_tokens": 17389614.0,
      "step": 10020
    },
    {
      "entropy": 5.6176183223724365,
      "epoch": 0.7800038902937172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004944626194946246,
      "loss": 5.414,
      "mean_token_accuracy": 0.1738124206662178,
      "num_tokens": 17398718.0,
      "step": 10025
    },
    {
      "entropy": 5.554498147964478,
      "epoch": 0.7803929196654348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944565068423945,
      "loss": 5.203,
      "mean_token_accuracy": 0.18256326615810395,
      "num_tokens": 17408504.0,
      "step": 10030
    },
    {
      "entropy": 5.4566785335540775,
      "epoch": 0.7807819490371523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004944503908602612,
      "loss": 5.1594,
      "mean_token_accuracy": 0.18209146857261657,
      "num_tokens": 17417826.0,
      "step": 10035
    },
    {
      "entropy": 5.590363836288452,
      "epoch": 0.7811709784088698,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944442715483174,
      "loss": 5.3467,
      "mean_token_accuracy": 0.17133993804454803,
      "num_tokens": 17426581.0,
      "step": 10040
    },
    {
      "entropy": 5.6116899967193605,
      "epoch": 0.7815600077805874,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944381489066561,
      "loss": 5.3081,
      "mean_token_accuracy": 0.17582398653030396,
      "num_tokens": 17435320.0,
      "step": 10045
    },
    {
      "entropy": 5.534014320373535,
      "epoch": 0.781949037152305,
      "grad_norm": 1.0,
      "learning_rate": 0.0004944320229353702,
      "loss": 5.247,
      "mean_token_accuracy": 0.1847595140337944,
      "num_tokens": 17443707.0,
      "step": 10050
    },
    {
      "entropy": 5.437810373306275,
      "epoch": 0.7823380665240226,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944258936345526,
      "loss": 5.133,
      "mean_token_accuracy": 0.18285618871450424,
      "num_tokens": 17452410.0,
      "step": 10055
    },
    {
      "entropy": 5.519442367553711,
      "epoch": 0.7827270958957401,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004944197610042963,
      "loss": 5.224,
      "mean_token_accuracy": 0.1846740022301674,
      "num_tokens": 17460000.0,
      "step": 10060
    },
    {
      "entropy": 5.542432737350464,
      "epoch": 0.7831161252674577,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004944136250446944,
      "loss": 5.2107,
      "mean_token_accuracy": 0.18191176801919937,
      "num_tokens": 17469179.0,
      "step": 10065
    },
    {
      "entropy": 5.5194908618927006,
      "epoch": 0.7835051546391752,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004944074857558399,
      "loss": 5.2328,
      "mean_token_accuracy": 0.1840083658695221,
      "num_tokens": 17478048.0,
      "step": 10070
    },
    {
      "entropy": 5.458236980438232,
      "epoch": 0.7838941840108928,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944013431378261,
      "loss": 5.131,
      "mean_token_accuracy": 0.18433062583208085,
      "num_tokens": 17485959.0,
      "step": 10075
    },
    {
      "entropy": 5.512459135055542,
      "epoch": 0.7842832133826104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943951971907461,
      "loss": 5.222,
      "mean_token_accuracy": 0.1845149204134941,
      "num_tokens": 17495158.0,
      "step": 10080
    },
    {
      "entropy": 5.522275066375732,
      "epoch": 0.7846722427543279,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943890479146933,
      "loss": 5.2476,
      "mean_token_accuracy": 0.18333764970302582,
      "num_tokens": 17503939.0,
      "step": 10085
    },
    {
      "entropy": 5.52121205329895,
      "epoch": 0.7850612721260455,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004943828953097607,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17532044351100923,
      "num_tokens": 17513391.0,
      "step": 10090
    },
    {
      "entropy": 5.599463081359863,
      "epoch": 0.7854503014977631,
      "grad_norm": 1.0,
      "learning_rate": 0.000494376739376042,
      "loss": 5.3201,
      "mean_token_accuracy": 0.17570871859788895,
      "num_tokens": 17522922.0,
      "step": 10095
    },
    {
      "entropy": 5.580346012115479,
      "epoch": 0.7858393308694807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004943705801136304,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17581601142883302,
      "num_tokens": 17531920.0,
      "step": 10100
    },
    {
      "entropy": 5.646449947357178,
      "epoch": 0.7862283602411982,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004943644175226193,
      "loss": 5.3353,
      "mean_token_accuracy": 0.1789740428328514,
      "num_tokens": 17539816.0,
      "step": 10105
    },
    {
      "entropy": 5.647852277755737,
      "epoch": 0.7866173896129158,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004943582516031025,
      "loss": 5.2833,
      "mean_token_accuracy": 0.18268545418977739,
      "num_tokens": 17548584.0,
      "step": 10110
    },
    {
      "entropy": 5.513976001739502,
      "epoch": 0.7870064189846333,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004943520823551732,
      "loss": 5.2978,
      "mean_token_accuracy": 0.18332510590553283,
      "num_tokens": 17557095.0,
      "step": 10115
    },
    {
      "entropy": 5.569630289077759,
      "epoch": 0.7873954483563509,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004943459097789252,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18324509412050247,
      "num_tokens": 17565937.0,
      "step": 10120
    },
    {
      "entropy": 5.516526746749878,
      "epoch": 0.7877844777280685,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004943397338744523,
      "loss": 5.3523,
      "mean_token_accuracy": 0.17756358534097672,
      "num_tokens": 17574948.0,
      "step": 10125
    },
    {
      "entropy": 5.619197940826416,
      "epoch": 0.7881735070997861,
      "grad_norm": 0.90625,
      "learning_rate": 0.000494333554641848,
      "loss": 5.3654,
      "mean_token_accuracy": 0.1765134796500206,
      "num_tokens": 17584307.0,
      "step": 10130
    },
    {
      "entropy": 5.545857238769531,
      "epoch": 0.7885625364715037,
      "grad_norm": 1.046875,
      "learning_rate": 0.000494327372081206,
      "loss": 5.2297,
      "mean_token_accuracy": 0.1844295084476471,
      "num_tokens": 17592828.0,
      "step": 10135
    },
    {
      "entropy": 5.581251811981201,
      "epoch": 0.7889515658432211,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004943211861926204,
      "loss": 5.2854,
      "mean_token_accuracy": 0.18676233440637588,
      "num_tokens": 17601110.0,
      "step": 10140
    },
    {
      "entropy": 5.5625608444213865,
      "epoch": 0.7893405952149387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943149969761848,
      "loss": 5.2436,
      "mean_token_accuracy": 0.18348842412233352,
      "num_tokens": 17610221.0,
      "step": 10145
    },
    {
      "entropy": 5.540854406356812,
      "epoch": 0.7897296245866563,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004943088044319932,
      "loss": 5.174,
      "mean_token_accuracy": 0.18292491137981415,
      "num_tokens": 17618343.0,
      "step": 10150
    },
    {
      "entropy": 5.546886253356933,
      "epoch": 0.7901186539583739,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943026085601395,
      "loss": 5.2839,
      "mean_token_accuracy": 0.1819344252347946,
      "num_tokens": 17626885.0,
      "step": 10155
    },
    {
      "entropy": 5.536163425445556,
      "epoch": 0.7905076833300915,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004942964093607178,
      "loss": 5.36,
      "mean_token_accuracy": 0.17669307738542556,
      "num_tokens": 17635891.0,
      "step": 10160
    },
    {
      "entropy": 5.57954740524292,
      "epoch": 0.790896712701809,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004942902068338222,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18569800853729249,
      "num_tokens": 17643815.0,
      "step": 10165
    },
    {
      "entropy": 5.565684032440186,
      "epoch": 0.7912857420735265,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004942840009795466,
      "loss": 5.2398,
      "mean_token_accuracy": 0.1842930570244789,
      "num_tokens": 17652414.0,
      "step": 10170
    },
    {
      "entropy": 5.547009086608886,
      "epoch": 0.7916747714452441,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942777917979855,
      "loss": 5.3034,
      "mean_token_accuracy": 0.17551807165145875,
      "num_tokens": 17661382.0,
      "step": 10175
    },
    {
      "entropy": 5.531071758270263,
      "epoch": 0.7920638008169617,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004942715792892328,
      "loss": 5.252,
      "mean_token_accuracy": 0.17437173128128053,
      "num_tokens": 17669745.0,
      "step": 10180
    },
    {
      "entropy": 5.560464191436767,
      "epoch": 0.7924528301886793,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000494265363453383,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18362361043691636,
      "num_tokens": 17678646.0,
      "step": 10185
    },
    {
      "entropy": 5.574765539169311,
      "epoch": 0.7928418595603968,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942591442905302,
      "loss": 5.342,
      "mean_token_accuracy": 0.17223089784383774,
      "num_tokens": 17687197.0,
      "step": 10190
    },
    {
      "entropy": 5.568060207366943,
      "epoch": 0.7932308889321144,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942529218007689,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17627218663692473,
      "num_tokens": 17696528.0,
      "step": 10195
    },
    {
      "entropy": 5.532153463363647,
      "epoch": 0.7936199183038319,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942466959841936,
      "loss": 5.2109,
      "mean_token_accuracy": 0.18019879162311553,
      "num_tokens": 17704688.0,
      "step": 10200
    },
    {
      "entropy": 5.437143087387085,
      "epoch": 0.7940089476755495,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004942404668408985,
      "loss": 5.1121,
      "mean_token_accuracy": 0.197529399394989,
      "num_tokens": 17713039.0,
      "step": 10205
    },
    {
      "entropy": 5.585537385940552,
      "epoch": 0.7943979770472671,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942342343709783,
      "loss": 5.4013,
      "mean_token_accuracy": 0.17179318517446518,
      "num_tokens": 17721771.0,
      "step": 10210
    },
    {
      "entropy": 5.5868367671966555,
      "epoch": 0.7947870064189846,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942279985745275,
      "loss": 5.2341,
      "mean_token_accuracy": 0.1795405477285385,
      "num_tokens": 17729982.0,
      "step": 10215
    },
    {
      "entropy": 5.548766279220581,
      "epoch": 0.7951760357907022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942217594516409,
      "loss": 5.2484,
      "mean_token_accuracy": 0.1787489965558052,
      "num_tokens": 17738397.0,
      "step": 10220
    },
    {
      "entropy": 5.618678569793701,
      "epoch": 0.7955650651624198,
      "grad_norm": 1.078125,
      "learning_rate": 0.000494215517002413,
      "loss": 5.359,
      "mean_token_accuracy": 0.17360880672931672,
      "num_tokens": 17747421.0,
      "step": 10225
    },
    {
      "entropy": 5.576296138763428,
      "epoch": 0.7959540945341373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004942092712269384,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17363293170928956,
      "num_tokens": 17756629.0,
      "step": 10230
    },
    {
      "entropy": 5.558225202560425,
      "epoch": 0.7963431239058549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004942030221253122,
      "loss": 5.2961,
      "mean_token_accuracy": 0.18580178320407867,
      "num_tokens": 17764595.0,
      "step": 10235
    },
    {
      "entropy": 5.547394037246704,
      "epoch": 0.7967321532775724,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004941967696976289,
      "loss": 5.182,
      "mean_token_accuracy": 0.18426274955272676,
      "num_tokens": 17773522.0,
      "step": 10240
    },
    {
      "entropy": 5.536650896072388,
      "epoch": 0.79712118264929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004941905139439835,
      "loss": 5.3027,
      "mean_token_accuracy": 0.183289997279644,
      "num_tokens": 17782850.0,
      "step": 10245
    },
    {
      "entropy": 5.470347118377686,
      "epoch": 0.7975102120210076,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004941842548644712,
      "loss": 5.158,
      "mean_token_accuracy": 0.18935055434703826,
      "num_tokens": 17790638.0,
      "step": 10250
    },
    {
      "entropy": 5.573339891433716,
      "epoch": 0.7978992413927252,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004941779924591864,
      "loss": 5.3018,
      "mean_token_accuracy": 0.17207662612199784,
      "num_tokens": 17799992.0,
      "step": 10255
    },
    {
      "entropy": 5.505545663833618,
      "epoch": 0.7982882707644428,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941717267282244,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1826378509402275,
      "num_tokens": 17808643.0,
      "step": 10260
    },
    {
      "entropy": 5.541860151290893,
      "epoch": 0.7986773001361602,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941654576716806,
      "loss": 5.2106,
      "mean_token_accuracy": 0.18334080129861832,
      "num_tokens": 17816551.0,
      "step": 10265
    },
    {
      "entropy": 5.516403532028198,
      "epoch": 0.7990663295078778,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941591852896495,
      "loss": 5.2391,
      "mean_token_accuracy": 0.18402206003665925,
      "num_tokens": 17825692.0,
      "step": 10270
    },
    {
      "entropy": 5.572368001937866,
      "epoch": 0.7994553588795954,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004941529095822268,
      "loss": 5.3479,
      "mean_token_accuracy": 0.17121150195598603,
      "num_tokens": 17833920.0,
      "step": 10275
    },
    {
      "entropy": 5.572353267669678,
      "epoch": 0.799844388251313,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004941466305495074,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17272062599658966,
      "num_tokens": 17842339.0,
      "step": 10280
    },
    {
      "entropy": 5.507529592514038,
      "epoch": 0.8002334176230306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941403481915867,
      "loss": 5.221,
      "mean_token_accuracy": 0.18210890889167786,
      "num_tokens": 17851273.0,
      "step": 10285
    },
    {
      "entropy": 5.511989450454712,
      "epoch": 0.800622446994748,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941340625085601,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18199315816164016,
      "num_tokens": 17860131.0,
      "step": 10290
    },
    {
      "entropy": 5.451030826568603,
      "epoch": 0.8010114763664656,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941277735005228,
      "loss": 5.183,
      "mean_token_accuracy": 0.18900831192731857,
      "num_tokens": 17868670.0,
      "step": 10295
    },
    {
      "entropy": 5.55652003288269,
      "epoch": 0.8014005057381832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941214811675702,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18667263090610503,
      "num_tokens": 17876972.0,
      "step": 10300
    },
    {
      "entropy": 5.5093494892120365,
      "epoch": 0.8017895351099008,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941151855097982,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17642293125391006,
      "num_tokens": 17885708.0,
      "step": 10305
    },
    {
      "entropy": 5.601290845870972,
      "epoch": 0.8021785644816184,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941088865273018,
      "loss": 5.291,
      "mean_token_accuracy": 0.17429496645927428,
      "num_tokens": 17894003.0,
      "step": 10310
    },
    {
      "entropy": 5.5281816005706785,
      "epoch": 0.802567593853336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004941025842201769,
      "loss": 5.1275,
      "mean_token_accuracy": 0.19140205681324005,
      "num_tokens": 17903113.0,
      "step": 10315
    },
    {
      "entropy": 5.514891910552978,
      "epoch": 0.8029566232250535,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940962785885189,
      "loss": 5.2165,
      "mean_token_accuracy": 0.17935004234313964,
      "num_tokens": 17912222.0,
      "step": 10320
    },
    {
      "entropy": 5.55033392906189,
      "epoch": 0.803345652596771,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940899696324237,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17741072028875352,
      "num_tokens": 17920900.0,
      "step": 10325
    },
    {
      "entropy": 5.60826849937439,
      "epoch": 0.8037346819684886,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940836573519868,
      "loss": 5.3233,
      "mean_token_accuracy": 0.17978237718343734,
      "num_tokens": 17929221.0,
      "step": 10330
    },
    {
      "entropy": 5.515179872512817,
      "epoch": 0.8041237113402062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004940773417473043,
      "loss": 5.235,
      "mean_token_accuracy": 0.18211790919303894,
      "num_tokens": 17938019.0,
      "step": 10335
    },
    {
      "entropy": 5.500888681411743,
      "epoch": 0.8045127407119238,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940710228184717,
      "loss": 5.2083,
      "mean_token_accuracy": 0.18903656899929047,
      "num_tokens": 17946972.0,
      "step": 10340
    },
    {
      "entropy": 5.590160894393921,
      "epoch": 0.8049017700836413,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494064700565585,
      "loss": 5.3162,
      "mean_token_accuracy": 0.17665481865406035,
      "num_tokens": 17956487.0,
      "step": 10345
    },
    {
      "entropy": 5.478394937515259,
      "epoch": 0.8052907994553589,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004940583749887403,
      "loss": 5.2167,
      "mean_token_accuracy": 0.18098169267177583,
      "num_tokens": 17965838.0,
      "step": 10350
    },
    {
      "entropy": 5.621855020523071,
      "epoch": 0.8056798288270764,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940520460880333,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17822395712137223,
      "num_tokens": 17974492.0,
      "step": 10355
    },
    {
      "entropy": 5.5938249111175535,
      "epoch": 0.806068858198794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940457138635601,
      "loss": 5.2904,
      "mean_token_accuracy": 0.18426005989313127,
      "num_tokens": 17983056.0,
      "step": 10360
    },
    {
      "entropy": 5.501426029205322,
      "epoch": 0.8064578875705116,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004940393783154169,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1829010561108589,
      "num_tokens": 17990822.0,
      "step": 10365
    },
    {
      "entropy": 5.47200608253479,
      "epoch": 0.8068469169422291,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004940330394436999,
      "loss": 5.1643,
      "mean_token_accuracy": 0.19085012078285218,
      "num_tokens": 17999284.0,
      "step": 10370
    },
    {
      "entropy": 5.548887825012207,
      "epoch": 0.8072359463139467,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000494026697248505,
      "loss": 5.3193,
      "mean_token_accuracy": 0.18064596951007844,
      "num_tokens": 18007585.0,
      "step": 10375
    },
    {
      "entropy": 5.448264741897583,
      "epoch": 0.8076249756856643,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004940203517299287,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18118565976619722,
      "num_tokens": 18016332.0,
      "step": 10380
    },
    {
      "entropy": 5.569855976104736,
      "epoch": 0.8080140050573819,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494014002888067,
      "loss": 5.2932,
      "mean_token_accuracy": 0.1763343408703804,
      "num_tokens": 18025524.0,
      "step": 10385
    },
    {
      "entropy": 5.552229404449463,
      "epoch": 0.8084030344290994,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940076507230166,
      "loss": 5.2792,
      "mean_token_accuracy": 0.1820867657661438,
      "num_tokens": 18033951.0,
      "step": 10390
    },
    {
      "entropy": 5.680067157745361,
      "epoch": 0.8087920638008169,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004940012952348735,
      "loss": 5.3969,
      "mean_token_accuracy": 0.17616181969642639,
      "num_tokens": 18042258.0,
      "step": 10395
    },
    {
      "entropy": 5.593752193450928,
      "epoch": 0.8091810931725345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004939949364237345,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18507616072893143,
      "num_tokens": 18050630.0,
      "step": 10400
    },
    {
      "entropy": 5.519416952133179,
      "epoch": 0.8095701225442521,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004939885742896958,
      "loss": 5.2703,
      "mean_token_accuracy": 0.18009306788444518,
      "num_tokens": 18059141.0,
      "step": 10405
    },
    {
      "entropy": 5.5265296459197994,
      "epoch": 0.8099591519159697,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493982208832854,
      "loss": 5.2749,
      "mean_token_accuracy": 0.17932132631540298,
      "num_tokens": 18067831.0,
      "step": 10410
    },
    {
      "entropy": 5.637141418457031,
      "epoch": 0.8103481812876873,
      "grad_norm": 1.125,
      "learning_rate": 0.0004939758400533058,
      "loss": 5.3654,
      "mean_token_accuracy": 0.17379961013793946,
      "num_tokens": 18076629.0,
      "step": 10415
    },
    {
      "entropy": 5.549745702743531,
      "epoch": 0.8107372106594047,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939694679511478,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18306296020746232,
      "num_tokens": 18085930.0,
      "step": 10420
    },
    {
      "entropy": 5.551803255081177,
      "epoch": 0.8111262400311223,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939630925264765,
      "loss": 5.2515,
      "mean_token_accuracy": 0.18021342754364014,
      "num_tokens": 18094947.0,
      "step": 10425
    },
    {
      "entropy": 5.555429410934448,
      "epoch": 0.8115152694028399,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000493956713779389,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1799653187394142,
      "num_tokens": 18103838.0,
      "step": 10430
    },
    {
      "entropy": 5.554565286636352,
      "epoch": 0.8119042987745575,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939503317099817,
      "loss": 5.177,
      "mean_token_accuracy": 0.1844314604997635,
      "num_tokens": 18112587.0,
      "step": 10435
    },
    {
      "entropy": 5.527368497848511,
      "epoch": 0.8122933281462751,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004939439463183517,
      "loss": 5.1996,
      "mean_token_accuracy": 0.18574080765247344,
      "num_tokens": 18121094.0,
      "step": 10440
    },
    {
      "entropy": 5.594413471221924,
      "epoch": 0.8126823575179926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004939375576045958,
      "loss": 5.3006,
      "mean_token_accuracy": 0.18318731486797332,
      "num_tokens": 18129699.0,
      "step": 10445
    },
    {
      "entropy": 5.570493602752686,
      "epoch": 0.8130713868897101,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004939311655688109,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1860767811536789,
      "num_tokens": 18138466.0,
      "step": 10450
    },
    {
      "entropy": 5.527794408798218,
      "epoch": 0.8134604162614277,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004939247702110941,
      "loss": 5.3087,
      "mean_token_accuracy": 0.1832980751991272,
      "num_tokens": 18147436.0,
      "step": 10455
    },
    {
      "entropy": 5.56290225982666,
      "epoch": 0.8138494456331453,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939183715315423,
      "loss": 5.2612,
      "mean_token_accuracy": 0.18561258018016816,
      "num_tokens": 18156228.0,
      "step": 10460
    },
    {
      "entropy": 5.56533784866333,
      "epoch": 0.8142384750048629,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004939119695302526,
      "loss": 5.2736,
      "mean_token_accuracy": 0.17960922569036483,
      "num_tokens": 18164571.0,
      "step": 10465
    },
    {
      "entropy": 5.551176643371582,
      "epoch": 0.8146275043765804,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004939055642073224,
      "loss": 5.2899,
      "mean_token_accuracy": 0.17826141268014908,
      "num_tokens": 18173322.0,
      "step": 10470
    },
    {
      "entropy": 5.590441417694092,
      "epoch": 0.815016533748298,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938991555628484,
      "loss": 5.3691,
      "mean_token_accuracy": 0.17837988436222077,
      "num_tokens": 18182306.0,
      "step": 10475
    },
    {
      "entropy": 5.642142868041992,
      "epoch": 0.8154055631200156,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938927435969283,
      "loss": 5.3275,
      "mean_token_accuracy": 0.17471859157085418,
      "num_tokens": 18190567.0,
      "step": 10480
    },
    {
      "entropy": 5.636589860916137,
      "epoch": 0.8157945924917331,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004938863283096592,
      "loss": 5.358,
      "mean_token_accuracy": 0.1726572796702385,
      "num_tokens": 18199678.0,
      "step": 10485
    },
    {
      "entropy": 5.484126138687134,
      "epoch": 0.8161836218634507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938799097011384,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18642169684171678,
      "num_tokens": 18208350.0,
      "step": 10490
    },
    {
      "entropy": 5.579818677902222,
      "epoch": 0.8165726512351682,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938734877714634,
      "loss": 5.2205,
      "mean_token_accuracy": 0.17800672203302384,
      "num_tokens": 18216618.0,
      "step": 10495
    },
    {
      "entropy": 5.56155366897583,
      "epoch": 0.8169616806068858,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938670625207317,
      "loss": 5.2438,
      "mean_token_accuracy": 0.18664785325527192,
      "num_tokens": 18224633.0,
      "step": 10500
    },
    {
      "entropy": 5.540253305435181,
      "epoch": 0.8173507099786034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004938606339490406,
      "loss": 5.1886,
      "mean_token_accuracy": 0.18040821105241775,
      "num_tokens": 18233212.0,
      "step": 10505
    },
    {
      "entropy": 5.608698081970215,
      "epoch": 0.817739739350321,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938542020564877,
      "loss": 5.3636,
      "mean_token_accuracy": 0.17857760190963745,
      "num_tokens": 18241796.0,
      "step": 10510
    },
    {
      "entropy": 5.5558243751525875,
      "epoch": 0.8181287687220385,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004938477668431706,
      "loss": 5.3693,
      "mean_token_accuracy": 0.1828357234597206,
      "num_tokens": 18251416.0,
      "step": 10515
    },
    {
      "entropy": 5.548902797698974,
      "epoch": 0.8185177980937561,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004938413283091871,
      "loss": 5.1576,
      "mean_token_accuracy": 0.18625995218753816,
      "num_tokens": 18259880.0,
      "step": 10520
    },
    {
      "entropy": 5.478700160980225,
      "epoch": 0.8189068274654736,
      "grad_norm": 1.125,
      "learning_rate": 0.0004938348864546347,
      "loss": 5.2436,
      "mean_token_accuracy": 0.18840804994106292,
      "num_tokens": 18268217.0,
      "step": 10525
    },
    {
      "entropy": 5.490329313278198,
      "epoch": 0.8192958568371912,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938284412796111,
      "loss": 5.301,
      "mean_token_accuracy": 0.18139395713806153,
      "num_tokens": 18276556.0,
      "step": 10530
    },
    {
      "entropy": 5.542969799041748,
      "epoch": 0.8196848862089088,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004938219927842144,
      "loss": 5.2535,
      "mean_token_accuracy": 0.18201657980680466,
      "num_tokens": 18285199.0,
      "step": 10535
    },
    {
      "entropy": 5.543492555618286,
      "epoch": 0.8200739155806264,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004938155409685422,
      "loss": 5.221,
      "mean_token_accuracy": 0.18483659029006957,
      "num_tokens": 18294004.0,
      "step": 10540
    },
    {
      "entropy": 5.479832887649536,
      "epoch": 0.820462944952344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938090858326923,
      "loss": 5.1451,
      "mean_token_accuracy": 0.1871393457055092,
      "num_tokens": 18301761.0,
      "step": 10545
    },
    {
      "entropy": 5.584003448486328,
      "epoch": 0.8208519743240614,
      "grad_norm": 1.0625,
      "learning_rate": 0.000493802627376763,
      "loss": 5.349,
      "mean_token_accuracy": 0.17411619573831558,
      "num_tokens": 18309555.0,
      "step": 10550
    },
    {
      "entropy": 5.5634064197540285,
      "epoch": 0.821241003695779,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937961656008518,
      "loss": 5.3124,
      "mean_token_accuracy": 0.17790265679359435,
      "num_tokens": 18318011.0,
      "step": 10555
    },
    {
      "entropy": 5.587578916549683,
      "epoch": 0.8216300330674966,
      "grad_norm": 1.0,
      "learning_rate": 0.0004937897005050573,
      "loss": 5.3019,
      "mean_token_accuracy": 0.17786704897880554,
      "num_tokens": 18326108.0,
      "step": 10560
    },
    {
      "entropy": 5.5615644454956055,
      "epoch": 0.8220190624392142,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.2564,
      "mean_token_accuracy": 0.1778743878006935,
      "num_tokens": 18334777.0,
      "step": 10565
    },
    {
      "entropy": 5.5428698539733885,
      "epoch": 0.8224080918109318,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004937767603542098,
      "loss": 5.223,
      "mean_token_accuracy": 0.17921229600906372,
      "num_tokens": 18343796.0,
      "step": 10570
    },
    {
      "entropy": 5.549758958816528,
      "epoch": 0.8227971211826492,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004937702852993534,
      "loss": 5.2602,
      "mean_token_accuracy": 0.1769055426120758,
      "num_tokens": 18352396.0,
      "step": 10575
    },
    {
      "entropy": 5.534566402435303,
      "epoch": 0.8231861505543668,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000493763806925006,
      "loss": 5.2718,
      "mean_token_accuracy": 0.1870693102478981,
      "num_tokens": 18360914.0,
      "step": 10580
    },
    {
      "entropy": 5.5200659275054935,
      "epoch": 0.8235751799260844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937573252312661,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18408682346343994,
      "num_tokens": 18369281.0,
      "step": 10585
    },
    {
      "entropy": 5.510223770141602,
      "epoch": 0.823964209297802,
      "grad_norm": 1.109375,
      "learning_rate": 0.000493750840218232,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19342626482248307,
      "num_tokens": 18377305.0,
      "step": 10590
    },
    {
      "entropy": 5.5342694282531735,
      "epoch": 0.8243532386695196,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004937443518860021,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1735801115632057,
      "num_tokens": 18386563.0,
      "step": 10595
    },
    {
      "entropy": 5.528779554367065,
      "epoch": 0.8247422680412371,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004937378602346747,
      "loss": 5.2364,
      "mean_token_accuracy": 0.1893967166543007,
      "num_tokens": 18395150.0,
      "step": 10600
    },
    {
      "entropy": 5.559446954727173,
      "epoch": 0.8251312974129547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004937313652643485,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17537704557180406,
      "num_tokens": 18403790.0,
      "step": 10605
    },
    {
      "entropy": 5.588216972351074,
      "epoch": 0.8255203267846722,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493724866975122,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18539801836013795,
      "num_tokens": 18411936.0,
      "step": 10610
    },
    {
      "entropy": 5.4975566387176515,
      "epoch": 0.8259093561563898,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937183653670937,
      "loss": 5.1205,
      "mean_token_accuracy": 0.1907249301671982,
      "num_tokens": 18421146.0,
      "step": 10615
    },
    {
      "entropy": 5.54138445854187,
      "epoch": 0.8262983855281074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937118604403623,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18405063152313234,
      "num_tokens": 18429498.0,
      "step": 10620
    },
    {
      "entropy": 5.594035053253174,
      "epoch": 0.8266874148998249,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004937053521950266,
      "loss": 5.3016,
      "mean_token_accuracy": 0.1795467182993889,
      "num_tokens": 18438600.0,
      "step": 10625
    },
    {
      "entropy": 5.529293155670166,
      "epoch": 0.8270764442715425,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004936988406311854,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18156871795654297,
      "num_tokens": 18447397.0,
      "step": 10630
    },
    {
      "entropy": 5.523371791839599,
      "epoch": 0.8274654736432601,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936923257489373,
      "loss": 5.2629,
      "mean_token_accuracy": 0.17718516588211058,
      "num_tokens": 18456677.0,
      "step": 10635
    },
    {
      "entropy": 5.508723735809326,
      "epoch": 0.8278545030149776,
      "grad_norm": 1.0,
      "learning_rate": 0.0004936858075483811,
      "loss": 5.1952,
      "mean_token_accuracy": 0.18520419001579286,
      "num_tokens": 18465605.0,
      "step": 10640
    },
    {
      "entropy": 5.508184766769409,
      "epoch": 0.8282435323866952,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493679286029616,
      "loss": 5.2287,
      "mean_token_accuracy": 0.1800902873277664,
      "num_tokens": 18474419.0,
      "step": 10645
    },
    {
      "entropy": 5.64747462272644,
      "epoch": 0.8286325617584127,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004936727611927407,
      "loss": 5.354,
      "mean_token_accuracy": 0.17717713862657547,
      "num_tokens": 18482646.0,
      "step": 10650
    },
    {
      "entropy": 5.54002046585083,
      "epoch": 0.8290215911301303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936662330378546,
      "loss": 5.2057,
      "mean_token_accuracy": 0.1830547034740448,
      "num_tokens": 18490899.0,
      "step": 10655
    },
    {
      "entropy": 5.526831150054932,
      "epoch": 0.8294106205018479,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936597015650561,
      "loss": 5.2091,
      "mean_token_accuracy": 0.18813900351524354,
      "num_tokens": 18499238.0,
      "step": 10660
    },
    {
      "entropy": 5.520270872116089,
      "epoch": 0.8297996498735655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004936531667744448,
      "loss": 5.1793,
      "mean_token_accuracy": 0.17968463003635407,
      "num_tokens": 18507668.0,
      "step": 10665
    },
    {
      "entropy": 5.588473844528198,
      "epoch": 0.8301886792452831,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936466286661197,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18253855854272844,
      "num_tokens": 18516366.0,
      "step": 10670
    },
    {
      "entropy": 5.5107598304748535,
      "epoch": 0.8305777086170005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00049364008724018,
      "loss": 5.2216,
      "mean_token_accuracy": 0.18302127122879028,
      "num_tokens": 18525012.0,
      "step": 10675
    },
    {
      "entropy": 5.5812681198120115,
      "epoch": 0.8309667379887181,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493633542496725,
      "loss": 5.3062,
      "mean_token_accuracy": 0.18337736427783966,
      "num_tokens": 18533277.0,
      "step": 10680
    },
    {
      "entropy": 5.57138500213623,
      "epoch": 0.8313557673604357,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004936269944358539,
      "loss": 5.3108,
      "mean_token_accuracy": 0.17496419250965117,
      "num_tokens": 18541738.0,
      "step": 10685
    },
    {
      "entropy": 5.58394193649292,
      "epoch": 0.8317447967321533,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004936204430576664,
      "loss": 5.271,
      "mean_token_accuracy": 0.18091964572668076,
      "num_tokens": 18550305.0,
      "step": 10690
    },
    {
      "entropy": 5.544063138961792,
      "epoch": 0.8321338261038709,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936138883622614,
      "loss": 5.2318,
      "mean_token_accuracy": 0.19183179289102553,
      "num_tokens": 18558528.0,
      "step": 10695
    },
    {
      "entropy": 5.469418525695801,
      "epoch": 0.8325228554755884,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004936073303497387,
      "loss": 5.1722,
      "mean_token_accuracy": 0.18760383278131484,
      "num_tokens": 18567166.0,
      "step": 10700
    },
    {
      "entropy": 5.494637823104858,
      "epoch": 0.8329118848473059,
      "grad_norm": 1.0,
      "learning_rate": 0.0004936007690201976,
      "loss": 5.1935,
      "mean_token_accuracy": 0.18928715586662292,
      "num_tokens": 18575986.0,
      "step": 10705
    },
    {
      "entropy": 5.599856281280518,
      "epoch": 0.8333009142190235,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493594204373738,
      "loss": 5.2465,
      "mean_token_accuracy": 0.17514331936836242,
      "num_tokens": 18585446.0,
      "step": 10710
    },
    {
      "entropy": 5.553297996520996,
      "epoch": 0.8336899435907411,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935876364104591,
      "loss": 5.2774,
      "mean_token_accuracy": 0.18375732898712158,
      "num_tokens": 18594236.0,
      "step": 10715
    },
    {
      "entropy": 5.539151382446289,
      "epoch": 0.8340789729624587,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004935810651304608,
      "loss": 5.2727,
      "mean_token_accuracy": 0.1795378729701042,
      "num_tokens": 18603229.0,
      "step": 10720
    },
    {
      "entropy": 5.518246507644653,
      "epoch": 0.8344680023341763,
      "grad_norm": 1.0,
      "learning_rate": 0.0004935744905338427,
      "loss": 5.1297,
      "mean_token_accuracy": 0.19666348099708558,
      "num_tokens": 18611783.0,
      "step": 10725
    },
    {
      "entropy": 5.610187101364136,
      "epoch": 0.8348570317058938,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935679126207046,
      "loss": 5.3275,
      "mean_token_accuracy": 0.17979582101106645,
      "num_tokens": 18620901.0,
      "step": 10730
    },
    {
      "entropy": 5.523731851577759,
      "epoch": 0.8352460610776113,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004935613313911463,
      "loss": 5.2326,
      "mean_token_accuracy": 0.17777478843927383,
      "num_tokens": 18629447.0,
      "step": 10735
    },
    {
      "entropy": 5.581677579879761,
      "epoch": 0.8356350904493289,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004935547468452678,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18409804850816727,
      "num_tokens": 18637895.0,
      "step": 10740
    },
    {
      "entropy": 5.576303434371948,
      "epoch": 0.8360241198210465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004935481589831688,
      "loss": 5.3899,
      "mean_token_accuracy": 0.17418855279684067,
      "num_tokens": 18646417.0,
      "step": 10745
    },
    {
      "entropy": 5.575652599334717,
      "epoch": 0.8364131491927641,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004935415678049492,
      "loss": 5.3088,
      "mean_token_accuracy": 0.17884970754384993,
      "num_tokens": 18654814.0,
      "step": 10750
    },
    {
      "entropy": 5.595830678939819,
      "epoch": 0.8368021785644816,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004935349733107094,
      "loss": 5.2989,
      "mean_token_accuracy": 0.18440109640359878,
      "num_tokens": 18663520.0,
      "step": 10755
    },
    {
      "entropy": 5.545376777648926,
      "epoch": 0.8371912079361992,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493528375500549,
      "loss": 5.3275,
      "mean_token_accuracy": 0.1782543271780014,
      "num_tokens": 18672033.0,
      "step": 10760
    },
    {
      "entropy": 5.54251537322998,
      "epoch": 0.8375802373079168,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004935217743745685,
      "loss": 5.1263,
      "mean_token_accuracy": 0.191961570084095,
      "num_tokens": 18680690.0,
      "step": 10765
    },
    {
      "entropy": 5.522523355484009,
      "epoch": 0.8379692666796343,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004935151699328677,
      "loss": 5.2418,
      "mean_token_accuracy": 0.17692058384418488,
      "num_tokens": 18688683.0,
      "step": 10770
    },
    {
      "entropy": 5.484722566604614,
      "epoch": 0.8383582960513519,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935085621755471,
      "loss": 5.3234,
      "mean_token_accuracy": 0.17820271998643875,
      "num_tokens": 18697273.0,
      "step": 10775
    },
    {
      "entropy": 5.616714000701904,
      "epoch": 0.8387473254230694,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493501951102707,
      "loss": 5.2921,
      "mean_token_accuracy": 0.17552858144044875,
      "num_tokens": 18705076.0,
      "step": 10780
    },
    {
      "entropy": 5.537790441513062,
      "epoch": 0.839136354794787,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004934953367144474,
      "loss": 5.1677,
      "mean_token_accuracy": 0.1860572099685669,
      "num_tokens": 18713596.0,
      "step": 10785
    },
    {
      "entropy": 5.485022783279419,
      "epoch": 0.8395253841665046,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934887190108688,
      "loss": 5.1857,
      "mean_token_accuracy": 0.1858276531100273,
      "num_tokens": 18722611.0,
      "step": 10790
    },
    {
      "entropy": 5.609499931335449,
      "epoch": 0.8399144135382222,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934820979920719,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1708538606762886,
      "num_tokens": 18731126.0,
      "step": 10795
    },
    {
      "entropy": 5.534915781021118,
      "epoch": 0.8403034429099397,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004934754736581567,
      "loss": 5.2294,
      "mean_token_accuracy": 0.17858947664499283,
      "num_tokens": 18739815.0,
      "step": 10800
    },
    {
      "entropy": 5.478046035766601,
      "epoch": 0.8406924722816572,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000493468846009224,
      "loss": 5.1545,
      "mean_token_accuracy": 0.19134612679481505,
      "num_tokens": 18748603.0,
      "step": 10805
    },
    {
      "entropy": 5.507309770584106,
      "epoch": 0.8410815016533748,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004934622150453742,
      "loss": 5.2552,
      "mean_token_accuracy": 0.17772441059350969,
      "num_tokens": 18757819.0,
      "step": 10810
    },
    {
      "entropy": 5.602052545547485,
      "epoch": 0.8414705310250924,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000493455580766708,
      "loss": 5.266,
      "mean_token_accuracy": 0.17887115776538848,
      "num_tokens": 18766135.0,
      "step": 10815
    },
    {
      "entropy": 5.515768527984619,
      "epoch": 0.84185956039681,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934489431733262,
      "loss": 5.1591,
      "mean_token_accuracy": 0.18530394583940507,
      "num_tokens": 18775202.0,
      "step": 10820
    },
    {
      "entropy": 5.508689069747925,
      "epoch": 0.8422485897685276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934423022653293,
      "loss": 5.2298,
      "mean_token_accuracy": 0.18810921162366867,
      "num_tokens": 18783466.0,
      "step": 10825
    },
    {
      "entropy": 5.543803453445435,
      "epoch": 0.842637619140245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004934356580428182,
      "loss": 5.2305,
      "mean_token_accuracy": 0.1796533927321434,
      "num_tokens": 18791772.0,
      "step": 10830
    },
    {
      "entropy": 5.617976713180542,
      "epoch": 0.8430266485119626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004934290105058937,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17383247017860412,
      "num_tokens": 18801094.0,
      "step": 10835
    },
    {
      "entropy": 5.553902578353882,
      "epoch": 0.8434156778836802,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004934223596546565,
      "loss": 5.1615,
      "mean_token_accuracy": 0.18740533292293549,
      "num_tokens": 18809256.0,
      "step": 10840
    },
    {
      "entropy": 5.533125305175782,
      "epoch": 0.8438047072553978,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934157054892077,
      "loss": 5.3314,
      "mean_token_accuracy": 0.18046908229589462,
      "num_tokens": 18818109.0,
      "step": 10845
    },
    {
      "entropy": 5.549737119674683,
      "epoch": 0.8441937366271154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934090480096482,
      "loss": 5.203,
      "mean_token_accuracy": 0.1879824861884117,
      "num_tokens": 18826330.0,
      "step": 10850
    },
    {
      "entropy": 5.629470348358154,
      "epoch": 0.8445827659988329,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004934023872160791,
      "loss": 5.4019,
      "mean_token_accuracy": 0.17526203095912934,
      "num_tokens": 18835280.0,
      "step": 10855
    },
    {
      "entropy": 5.499673795700073,
      "epoch": 0.8449717953705504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004933957231086014,
      "loss": 5.1609,
      "mean_token_accuracy": 0.1841066911816597,
      "num_tokens": 18843300.0,
      "step": 10860
    },
    {
      "entropy": 5.463746643066406,
      "epoch": 0.845360824742268,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004933890556873161,
      "loss": 5.2548,
      "mean_token_accuracy": 0.17553244531154633,
      "num_tokens": 18852257.0,
      "step": 10865
    },
    {
      "entropy": 5.56828932762146,
      "epoch": 0.8457498541139856,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004933823849523246,
      "loss": 5.2856,
      "mean_token_accuracy": 0.18161431699991226,
      "num_tokens": 18860621.0,
      "step": 10870
    },
    {
      "entropy": 5.5673925399780275,
      "epoch": 0.8461388834857032,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933757109037279,
      "loss": 5.1902,
      "mean_token_accuracy": 0.18614138811826705,
      "num_tokens": 18869390.0,
      "step": 10875
    },
    {
      "entropy": 5.555185747146607,
      "epoch": 0.8465279128574207,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933690335416274,
      "loss": 5.2911,
      "mean_token_accuracy": 0.17937638312578202,
      "num_tokens": 18877942.0,
      "step": 10880
    },
    {
      "entropy": 5.503046321868896,
      "epoch": 0.8469169422291383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933623528661245,
      "loss": 5.2779,
      "mean_token_accuracy": 0.1784745156764984,
      "num_tokens": 18886752.0,
      "step": 10885
    },
    {
      "entropy": 5.5775063037872314,
      "epoch": 0.8473059716008559,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933556688773203,
      "loss": 5.2903,
      "mean_token_accuracy": 0.17868355214595794,
      "num_tokens": 18895708.0,
      "step": 10890
    },
    {
      "entropy": 5.567748641967773,
      "epoch": 0.8476950009725734,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933489815753165,
      "loss": 5.2268,
      "mean_token_accuracy": 0.18473885357379913,
      "num_tokens": 18904610.0,
      "step": 10895
    },
    {
      "entropy": 5.520393228530883,
      "epoch": 0.848084030344291,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933422909602143,
      "loss": 5.2711,
      "mean_token_accuracy": 0.1805527240037918,
      "num_tokens": 18912886.0,
      "step": 10900
    },
    {
      "entropy": 5.569060850143432,
      "epoch": 0.8484730597160085,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004933355970321155,
      "loss": 5.3132,
      "mean_token_accuracy": 0.17902422547340394,
      "num_tokens": 18921685.0,
      "step": 10905
    },
    {
      "entropy": 5.564014148712158,
      "epoch": 0.8488620890877261,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933288997911215,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1850571870803833,
      "num_tokens": 18929843.0,
      "step": 10910
    },
    {
      "entropy": 5.497473621368409,
      "epoch": 0.8492511184594437,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004933221992373338,
      "loss": 5.2166,
      "mean_token_accuracy": 0.17965936958789824,
      "num_tokens": 18938402.0,
      "step": 10915
    },
    {
      "entropy": 5.563475131988525,
      "epoch": 0.8496401478311613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004933154953708544,
      "loss": 5.3496,
      "mean_token_accuracy": 0.1703747108578682,
      "num_tokens": 18947643.0,
      "step": 10920
    },
    {
      "entropy": 5.616429376602173,
      "epoch": 0.8500291772028788,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004933087881917848,
      "loss": 5.2793,
      "mean_token_accuracy": 0.17879072725772857,
      "num_tokens": 18956320.0,
      "step": 10925
    },
    {
      "entropy": 5.681483268737793,
      "epoch": 0.8504182065745964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933020777002268,
      "loss": 5.2558,
      "mean_token_accuracy": 0.1776057004928589,
      "num_tokens": 18964804.0,
      "step": 10930
    },
    {
      "entropy": 5.464506435394287,
      "epoch": 0.8508072359463139,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004932953638962823,
      "loss": 5.2032,
      "mean_token_accuracy": 0.1821805253624916,
      "num_tokens": 18973061.0,
      "step": 10935
    },
    {
      "entropy": 5.506009244918824,
      "epoch": 0.8511962653180315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000493288646780053,
      "loss": 5.1873,
      "mean_token_accuracy": 0.18650328516960143,
      "num_tokens": 18981604.0,
      "step": 10940
    },
    {
      "entropy": 5.55337553024292,
      "epoch": 0.8515852946897491,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932819263516409,
      "loss": 5.2905,
      "mean_token_accuracy": 0.17652418315410615,
      "num_tokens": 18990194.0,
      "step": 10945
    },
    {
      "entropy": 5.532214689254761,
      "epoch": 0.8519743240614667,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932752026111481,
      "loss": 5.1958,
      "mean_token_accuracy": 0.181524658203125,
      "num_tokens": 18998803.0,
      "step": 10950
    },
    {
      "entropy": 5.625107955932617,
      "epoch": 0.8523633534331843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004932684755586765,
      "loss": 5.3102,
      "mean_token_accuracy": 0.17468704134225846,
      "num_tokens": 19007247.0,
      "step": 10955
    },
    {
      "entropy": 5.5398841381073,
      "epoch": 0.8527523828049017,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932617451943282,
      "loss": 5.233,
      "mean_token_accuracy": 0.18560332357883452,
      "num_tokens": 19015436.0,
      "step": 10960
    },
    {
      "entropy": 5.594821643829346,
      "epoch": 0.8531414121766193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004932550115182053,
      "loss": 5.329,
      "mean_token_accuracy": 0.17277351170778274,
      "num_tokens": 19023673.0,
      "step": 10965
    },
    {
      "entropy": 5.534565210342407,
      "epoch": 0.8535304415483369,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049324827453041,
      "loss": 5.1617,
      "mean_token_accuracy": 0.1841248795390129,
      "num_tokens": 19031714.0,
      "step": 10970
    },
    {
      "entropy": 5.432816839218139,
      "epoch": 0.8539194709200545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004932415342310446,
      "loss": 5.1636,
      "mean_token_accuracy": 0.1836773708462715,
      "num_tokens": 19039910.0,
      "step": 10975
    },
    {
      "entropy": 5.564401769638062,
      "epoch": 0.8543085002917721,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004932347906202111,
      "loss": 5.2955,
      "mean_token_accuracy": 0.18685130327939986,
      "num_tokens": 19048293.0,
      "step": 10980
    },
    {
      "entropy": 5.569058752059936,
      "epoch": 0.8546975296634896,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493228043698012,
      "loss": 5.2267,
      "mean_token_accuracy": 0.17633443474769592,
      "num_tokens": 19056209.0,
      "step": 10985
    },
    {
      "entropy": 5.5361498355865475,
      "epoch": 0.8550865590352071,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932212934645498,
      "loss": 5.131,
      "mean_token_accuracy": 0.18912424743175507,
      "num_tokens": 19063974.0,
      "step": 10990
    },
    {
      "entropy": 5.55227575302124,
      "epoch": 0.8554755884069247,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932145399199268,
      "loss": 5.3794,
      "mean_token_accuracy": 0.17500831484794616,
      "num_tokens": 19072531.0,
      "step": 10995
    },
    {
      "entropy": 5.653836059570312,
      "epoch": 0.8558646177786423,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004932077830642455,
      "loss": 5.3062,
      "mean_token_accuracy": 0.17791387736797332,
      "num_tokens": 19081299.0,
      "step": 11000
    },
    {
      "entropy": 5.5676939487457275,
      "epoch": 0.8562536471503599,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932010228976084,
      "loss": 5.2054,
      "mean_token_accuracy": 0.19290074110031127,
      "num_tokens": 19089599.0,
      "step": 11005
    },
    {
      "entropy": 5.505162048339844,
      "epoch": 0.8566426765220774,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000493194259420118,
      "loss": 5.1904,
      "mean_token_accuracy": 0.18730612546205522,
      "num_tokens": 19097910.0,
      "step": 11010
    },
    {
      "entropy": 5.47117166519165,
      "epoch": 0.857031705893795,
      "grad_norm": 1.09375,
      "learning_rate": 0.000493187492631877,
      "loss": 5.1448,
      "mean_token_accuracy": 0.19171023964881898,
      "num_tokens": 19105939.0,
      "step": 11015
    },
    {
      "entropy": 5.477901887893677,
      "epoch": 0.8574207352655125,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004931807225329882,
      "loss": 5.2035,
      "mean_token_accuracy": 0.18485299944877626,
      "num_tokens": 19115407.0,
      "step": 11020
    },
    {
      "entropy": 5.48896336555481,
      "epoch": 0.8578097646372301,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004931739491235541,
      "loss": 5.1254,
      "mean_token_accuracy": 0.19273362755775453,
      "num_tokens": 19123715.0,
      "step": 11025
    },
    {
      "entropy": 5.63070068359375,
      "epoch": 0.8581987940089477,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004931671724036777,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17230788320302964,
      "num_tokens": 19132978.0,
      "step": 11030
    },
    {
      "entropy": 5.544386148452759,
      "epoch": 0.8585878233806652,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931603923734616,
      "loss": 5.304,
      "mean_token_accuracy": 0.18017314970493317,
      "num_tokens": 19141822.0,
      "step": 11035
    },
    {
      "entropy": 5.48143048286438,
      "epoch": 0.8589768527523828,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004931536090330088,
      "loss": 5.1992,
      "mean_token_accuracy": 0.17714929580688477,
      "num_tokens": 19150612.0,
      "step": 11040
    },
    {
      "entropy": 5.528379964828491,
      "epoch": 0.8593658821241004,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931468223824222,
      "loss": 5.154,
      "mean_token_accuracy": 0.187422776222229,
      "num_tokens": 19159266.0,
      "step": 11045
    },
    {
      "entropy": 5.66276741027832,
      "epoch": 0.859754911495818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931400324218048,
      "loss": 5.3327,
      "mean_token_accuracy": 0.18183230310678483,
      "num_tokens": 19167111.0,
      "step": 11050
    },
    {
      "entropy": 5.5265590190887455,
      "epoch": 0.8601439408675355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004931332391512597,
      "loss": 5.299,
      "mean_token_accuracy": 0.17914700508117676,
      "num_tokens": 19177152.0,
      "step": 11055
    },
    {
      "entropy": 5.481976747512817,
      "epoch": 0.860532970239253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004931264425708897,
      "loss": 5.1277,
      "mean_token_accuracy": 0.18320668041706084,
      "num_tokens": 19185692.0,
      "step": 11060
    },
    {
      "entropy": 5.541192865371704,
      "epoch": 0.8609219996109706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931196426807983,
      "loss": 5.2936,
      "mean_token_accuracy": 0.18152756094932557,
      "num_tokens": 19194620.0,
      "step": 11065
    },
    {
      "entropy": 5.561440801620483,
      "epoch": 0.8613110289826882,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004931128394810884,
      "loss": 5.1748,
      "mean_token_accuracy": 0.18811147212982177,
      "num_tokens": 19202569.0,
      "step": 11070
    },
    {
      "entropy": 5.5706174850463865,
      "epoch": 0.8617000583544058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004931060329718634,
      "loss": 5.3196,
      "mean_token_accuracy": 0.1782485216856003,
      "num_tokens": 19211117.0,
      "step": 11075
    },
    {
      "entropy": 5.602764034271241,
      "epoch": 0.8620890877261234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004930992231532265,
      "loss": 5.3058,
      "mean_token_accuracy": 0.1797836884856224,
      "num_tokens": 19220015.0,
      "step": 11080
    },
    {
      "entropy": 5.619830846786499,
      "epoch": 0.8624781170978408,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000493092410025281,
      "loss": 5.3065,
      "mean_token_accuracy": 0.17686312794685363,
      "num_tokens": 19228344.0,
      "step": 11085
    },
    {
      "entropy": 5.50443320274353,
      "epoch": 0.8628671464695584,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004930855935881302,
      "loss": 5.1899,
      "mean_token_accuracy": 0.19230535328388215,
      "num_tokens": 19236776.0,
      "step": 11090
    },
    {
      "entropy": 5.623230743408203,
      "epoch": 0.863256175841276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930787738418777,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1671035945415497,
      "num_tokens": 19246107.0,
      "step": 11095
    },
    {
      "entropy": 5.607535982131958,
      "epoch": 0.8636452052129936,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930719507866269,
      "loss": 5.2901,
      "mean_token_accuracy": 0.18587073683738708,
      "num_tokens": 19254549.0,
      "step": 11100
    },
    {
      "entropy": 5.572094249725342,
      "epoch": 0.8640342345847112,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930651244224814,
      "loss": 5.2949,
      "mean_token_accuracy": 0.1801662638783455,
      "num_tokens": 19263024.0,
      "step": 11105
    },
    {
      "entropy": 5.512962055206299,
      "epoch": 0.8644232639564287,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004930582947495448,
      "loss": 5.208,
      "mean_token_accuracy": 0.18059032112360002,
      "num_tokens": 19271442.0,
      "step": 11110
    },
    {
      "entropy": 5.517531108856201,
      "epoch": 0.8648122933281462,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004930514617679206,
      "loss": 5.1614,
      "mean_token_accuracy": 0.19100626111030578,
      "num_tokens": 19279869.0,
      "step": 11115
    },
    {
      "entropy": 5.5686924934387205,
      "epoch": 0.8652013226998638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004930446254777125,
      "loss": 5.3574,
      "mean_token_accuracy": 0.17958442270755767,
      "num_tokens": 19288726.0,
      "step": 11120
    },
    {
      "entropy": 5.583649730682373,
      "epoch": 0.8655903520715814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930377858790242,
      "loss": 5.3107,
      "mean_token_accuracy": 0.1751727432012558,
      "num_tokens": 19297546.0,
      "step": 11125
    },
    {
      "entropy": 5.6224347114562985,
      "epoch": 0.865979381443299,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004930309429719595,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17556509673595427,
      "num_tokens": 19306060.0,
      "step": 11130
    },
    {
      "entropy": 5.566957283020019,
      "epoch": 0.8663684108150166,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930240967566224,
      "loss": 5.2567,
      "mean_token_accuracy": 0.18031541854143143,
      "num_tokens": 19314759.0,
      "step": 11135
    },
    {
      "entropy": 5.527809047698975,
      "epoch": 0.8667574401867341,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930172472331167,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18215182721614837,
      "num_tokens": 19322922.0,
      "step": 11140
    },
    {
      "entropy": 5.4821202754974365,
      "epoch": 0.8671464695584516,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004930103944015463,
      "loss": 5.1269,
      "mean_token_accuracy": 0.18408451080322266,
      "num_tokens": 19331946.0,
      "step": 11145
    },
    {
      "entropy": 5.600930023193359,
      "epoch": 0.8675354989301692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004930035382620149,
      "loss": 5.2751,
      "mean_token_accuracy": 0.1778065100312233,
      "num_tokens": 19340512.0,
      "step": 11150
    },
    {
      "entropy": 5.529328298568726,
      "epoch": 0.8679245283018868,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492996678814627,
      "loss": 5.2405,
      "mean_token_accuracy": 0.1811877131462097,
      "num_tokens": 19349135.0,
      "step": 11155
    },
    {
      "entropy": 5.535189580917359,
      "epoch": 0.8683135576736044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004929898160594865,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17958931028842925,
      "num_tokens": 19358593.0,
      "step": 11160
    },
    {
      "entropy": 5.6465692043304445,
      "epoch": 0.8687025870453219,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004929829499966974,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17606135904788972,
      "num_tokens": 19367366.0,
      "step": 11165
    },
    {
      "entropy": 5.592950296401978,
      "epoch": 0.8690916164170395,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004929760806263641,
      "loss": 5.3247,
      "mean_token_accuracy": 0.1734875425696373,
      "num_tokens": 19376872.0,
      "step": 11170
    },
    {
      "entropy": 5.615838670730591,
      "epoch": 0.8694806457887571,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929692079485906,
      "loss": 5.3295,
      "mean_token_accuracy": 0.17325059622526168,
      "num_tokens": 19385582.0,
      "step": 11175
    },
    {
      "entropy": 5.5427587032318115,
      "epoch": 0.8698696751604746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929623319634814,
      "loss": 5.2206,
      "mean_token_accuracy": 0.1821684569120407,
      "num_tokens": 19394102.0,
      "step": 11180
    },
    {
      "entropy": 5.562895441055298,
      "epoch": 0.8702587045321922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929554526711407,
      "loss": 5.2562,
      "mean_token_accuracy": 0.17677873075008393,
      "num_tokens": 19403170.0,
      "step": 11185
    },
    {
      "entropy": 5.5495288372039795,
      "epoch": 0.8706477339039097,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929485700716729,
      "loss": 5.2215,
      "mean_token_accuracy": 0.1873213306069374,
      "num_tokens": 19412396.0,
      "step": 11190
    },
    {
      "entropy": 5.550656509399414,
      "epoch": 0.8710367632756273,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004929416841651824,
      "loss": 5.2,
      "mean_token_accuracy": 0.18274609446525575,
      "num_tokens": 19421247.0,
      "step": 11195
    },
    {
      "entropy": 5.526799917221069,
      "epoch": 0.8714257926473449,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004929347949517739,
      "loss": 5.2357,
      "mean_token_accuracy": 0.1788181945681572,
      "num_tokens": 19429412.0,
      "step": 11200
    },
    {
      "entropy": 5.582285690307617,
      "epoch": 0.8718148220190625,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004929279024315516,
      "loss": 5.261,
      "mean_token_accuracy": 0.17433381527662278,
      "num_tokens": 19437551.0,
      "step": 11205
    },
    {
      "entropy": 5.462672519683838,
      "epoch": 0.87220385139078,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004929210066046204,
      "loss": 5.0975,
      "mean_token_accuracy": 0.18734758198261262,
      "num_tokens": 19446356.0,
      "step": 11210
    },
    {
      "entropy": 5.46998782157898,
      "epoch": 0.8725928807624975,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004929141074710847,
      "loss": 5.0587,
      "mean_token_accuracy": 0.19531500339508057,
      "num_tokens": 19454142.0,
      "step": 11215
    },
    {
      "entropy": 5.617652940750122,
      "epoch": 0.8729819101342151,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004929072050310491,
      "loss": 5.3743,
      "mean_token_accuracy": 0.1781470462679863,
      "num_tokens": 19462849.0,
      "step": 11220
    },
    {
      "entropy": 5.5333860397338865,
      "epoch": 0.8733709395059327,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929002992846188,
      "loss": 5.1635,
      "mean_token_accuracy": 0.18983460515737532,
      "num_tokens": 19471287.0,
      "step": 11225
    },
    {
      "entropy": 5.52715311050415,
      "epoch": 0.8737599688776503,
      "grad_norm": 1.0,
      "learning_rate": 0.0004928933902318981,
      "loss": 5.2262,
      "mean_token_accuracy": 0.18716436475515366,
      "num_tokens": 19480661.0,
      "step": 11230
    },
    {
      "entropy": 5.476283359527588,
      "epoch": 0.8741489982493679,
      "grad_norm": 1.125,
      "learning_rate": 0.000492886477872992,
      "loss": 5.1206,
      "mean_token_accuracy": 0.1878957837820053,
      "num_tokens": 19489716.0,
      "step": 11235
    },
    {
      "entropy": 5.516184234619141,
      "epoch": 0.8745380276210853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004928795622080054,
      "loss": 5.271,
      "mean_token_accuracy": 0.17724187672138214,
      "num_tokens": 19499019.0,
      "step": 11240
    },
    {
      "entropy": 5.5679398536682125,
      "epoch": 0.8749270569928029,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004928726432370434,
      "loss": 5.2028,
      "mean_token_accuracy": 0.18291952461004257,
      "num_tokens": 19507281.0,
      "step": 11245
    },
    {
      "entropy": 5.585212564468383,
      "epoch": 0.8753160863645205,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928657209602107,
      "loss": 5.2724,
      "mean_token_accuracy": 0.18346301317214966,
      "num_tokens": 19516196.0,
      "step": 11250
    },
    {
      "entropy": 5.568840551376343,
      "epoch": 0.8757051157362381,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004928587953776125,
      "loss": 5.301,
      "mean_token_accuracy": 0.1767169713973999,
      "num_tokens": 19523966.0,
      "step": 11255
    },
    {
      "entropy": 5.544528913497925,
      "epoch": 0.8760941451079557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928518664893538,
      "loss": 5.1989,
      "mean_token_accuracy": 0.18428248316049575,
      "num_tokens": 19532848.0,
      "step": 11260
    },
    {
      "entropy": 5.5038737773895265,
      "epoch": 0.8764831744796732,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928449342955397,
      "loss": 5.1579,
      "mean_token_accuracy": 0.18306698352098466,
      "num_tokens": 19540929.0,
      "step": 11265
    },
    {
      "entropy": 5.5189353942871096,
      "epoch": 0.8768722038513908,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928379987962756,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1850006937980652,
      "num_tokens": 19549628.0,
      "step": 11270
    },
    {
      "entropy": 5.520605087280273,
      "epoch": 0.8772612332231083,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004928310599916666,
      "loss": 5.2004,
      "mean_token_accuracy": 0.17930981069803237,
      "num_tokens": 19559002.0,
      "step": 11275
    },
    {
      "entropy": 5.522569465637207,
      "epoch": 0.8776502625948259,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928241178818178,
      "loss": 5.211,
      "mean_token_accuracy": 0.185836298763752,
      "num_tokens": 19567757.0,
      "step": 11280
    },
    {
      "entropy": 5.536190605163574,
      "epoch": 0.8780392919665435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928171724668349,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19180002361536025,
      "num_tokens": 19576595.0,
      "step": 11285
    },
    {
      "entropy": 5.529078960418701,
      "epoch": 0.878428321338261,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928102237468229,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17619390189647674,
      "num_tokens": 19585237.0,
      "step": 11290
    },
    {
      "entropy": 5.504305791854859,
      "epoch": 0.8788173507099786,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928032717218876,
      "loss": 5.1496,
      "mean_token_accuracy": 0.18973541706800462,
      "num_tokens": 19594474.0,
      "step": 11295
    },
    {
      "entropy": 5.515873193740845,
      "epoch": 0.8792063800816962,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927963163921343,
      "loss": 5.1986,
      "mean_token_accuracy": 0.18612133711576462,
      "num_tokens": 19603121.0,
      "step": 11300
    },
    {
      "entropy": 5.548409557342529,
      "epoch": 0.8795954094534137,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004927893577576685,
      "loss": 5.2416,
      "mean_token_accuracy": 0.178195983171463,
      "num_tokens": 19611820.0,
      "step": 11305
    },
    {
      "entropy": 5.574872350692749,
      "epoch": 0.8799844388251313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004927823958185959,
      "loss": 5.2413,
      "mean_token_accuracy": 0.1796655535697937,
      "num_tokens": 19620265.0,
      "step": 11310
    },
    {
      "entropy": 5.457470178604126,
      "epoch": 0.8803734681968488,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000492775430575022,
      "loss": 5.0919,
      "mean_token_accuracy": 0.18922037333250047,
      "num_tokens": 19628847.0,
      "step": 11315
    },
    {
      "entropy": 5.51951265335083,
      "epoch": 0.8807624975685664,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004927684620270527,
      "loss": 5.2556,
      "mean_token_accuracy": 0.18119800239801406,
      "num_tokens": 19637636.0,
      "step": 11320
    },
    {
      "entropy": 5.527203226089478,
      "epoch": 0.881151526940284,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004927614901747935,
      "loss": 5.2369,
      "mean_token_accuracy": 0.1773467481136322,
      "num_tokens": 19646663.0,
      "step": 11325
    },
    {
      "entropy": 5.531182050704956,
      "epoch": 0.8815405563120016,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004927545150183503,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18406083434820175,
      "num_tokens": 19655382.0,
      "step": 11330
    },
    {
      "entropy": 5.52825083732605,
      "epoch": 0.8819295856837192,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492747536557829,
      "loss": 5.2201,
      "mean_token_accuracy": 0.18831291794776917,
      "num_tokens": 19663851.0,
      "step": 11335
    },
    {
      "entropy": 5.479758596420288,
      "epoch": 0.8823186150554367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004927405547933353,
      "loss": 5.1738,
      "mean_token_accuracy": 0.1878650203347206,
      "num_tokens": 19672813.0,
      "step": 11340
    },
    {
      "entropy": 5.5052190780639645,
      "epoch": 0.8827076444271542,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004927335697249753,
      "loss": 5.1695,
      "mean_token_accuracy": 0.18782930672168732,
      "num_tokens": 19682240.0,
      "step": 11345
    },
    {
      "entropy": 5.569630765914917,
      "epoch": 0.8830966737988718,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927265813528549,
      "loss": 5.2212,
      "mean_token_accuracy": 0.18088840693235397,
      "num_tokens": 19691080.0,
      "step": 11350
    },
    {
      "entropy": 5.5099883556365965,
      "epoch": 0.8834857031705894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004927195896770804,
      "loss": 5.1779,
      "mean_token_accuracy": 0.1918042242527008,
      "num_tokens": 19700067.0,
      "step": 11355
    },
    {
      "entropy": 5.529403257369995,
      "epoch": 0.883874732542307,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004927125946977574,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18812153488397598,
      "num_tokens": 19708236.0,
      "step": 11360
    },
    {
      "entropy": 5.560075139999389,
      "epoch": 0.8842637619140246,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004927055964149923,
      "loss": 5.3031,
      "mean_token_accuracy": 0.18180075138807297,
      "num_tokens": 19717225.0,
      "step": 11365
    },
    {
      "entropy": 5.522982168197632,
      "epoch": 0.884652791285742,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004926985948288914,
      "loss": 5.2625,
      "mean_token_accuracy": 0.1891829937696457,
      "num_tokens": 19725169.0,
      "step": 11370
    },
    {
      "entropy": 5.536978340148925,
      "epoch": 0.8850418206574596,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004926915899395608,
      "loss": 5.2319,
      "mean_token_accuracy": 0.1837349683046341,
      "num_tokens": 19733937.0,
      "step": 11375
    },
    {
      "entropy": 5.510504961013794,
      "epoch": 0.8854308500291772,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004926845817471068,
      "loss": 5.1838,
      "mean_token_accuracy": 0.19284587651491164,
      "num_tokens": 19742661.0,
      "step": 11380
    },
    {
      "entropy": 5.487407159805298,
      "epoch": 0.8858198794008948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004926775702516358,
      "loss": 5.244,
      "mean_token_accuracy": 0.18677958250045776,
      "num_tokens": 19751314.0,
      "step": 11385
    },
    {
      "entropy": 5.523621511459351,
      "epoch": 0.8862089087726124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004926705554532541,
      "loss": 5.1774,
      "mean_token_accuracy": 0.1917124480009079,
      "num_tokens": 19760144.0,
      "step": 11390
    },
    {
      "entropy": 5.505693769454956,
      "epoch": 0.8865979381443299,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926635373520682,
      "loss": 5.1946,
      "mean_token_accuracy": 0.18730054050683975,
      "num_tokens": 19768574.0,
      "step": 11395
    },
    {
      "entropy": 5.585032606124878,
      "epoch": 0.8869869675160474,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926565159481845,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1737936854362488,
      "num_tokens": 19778383.0,
      "step": 11400
    },
    {
      "entropy": 5.551654720306397,
      "epoch": 0.887375996887765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926494912417097,
      "loss": 5.2116,
      "mean_token_accuracy": 0.1848152220249176,
      "num_tokens": 19787707.0,
      "step": 11405
    },
    {
      "entropy": 5.531039810180664,
      "epoch": 0.8877650262594826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004926424632327503,
      "loss": 5.1094,
      "mean_token_accuracy": 0.19663903564214708,
      "num_tokens": 19795909.0,
      "step": 11410
    },
    {
      "entropy": 5.4795300483703615,
      "epoch": 0.8881540556312002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004926354319214129,
      "loss": 5.1733,
      "mean_token_accuracy": 0.18454110771417617,
      "num_tokens": 19804094.0,
      "step": 11415
    },
    {
      "entropy": 5.460201930999756,
      "epoch": 0.8885430850029177,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004926283973078041,
      "loss": 5.0873,
      "mean_token_accuracy": 0.18923551142215728,
      "num_tokens": 19813061.0,
      "step": 11420
    },
    {
      "entropy": 5.472792053222657,
      "epoch": 0.8889321143746353,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004926213593920309,
      "loss": 5.101,
      "mean_token_accuracy": 0.1912500724196434,
      "num_tokens": 19821432.0,
      "step": 11425
    },
    {
      "entropy": 5.5045366287231445,
      "epoch": 0.8893211437463528,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926143181742,
      "loss": 5.2396,
      "mean_token_accuracy": 0.18476988971233368,
      "num_tokens": 19830278.0,
      "step": 11430
    },
    {
      "entropy": 5.429034233093262,
      "epoch": 0.8897101731180704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004926072736544181,
      "loss": 5.1641,
      "mean_token_accuracy": 0.18787843286991118,
      "num_tokens": 19838872.0,
      "step": 11435
    },
    {
      "entropy": 5.632330846786499,
      "epoch": 0.890099202489788,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004926002258327922,
      "loss": 5.2996,
      "mean_token_accuracy": 0.17796659767627715,
      "num_tokens": 19848032.0,
      "step": 11440
    },
    {
      "entropy": 5.509681034088135,
      "epoch": 0.8904882318615055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004925931747094292,
      "loss": 5.1606,
      "mean_token_accuracy": 0.19246817976236344,
      "num_tokens": 19856711.0,
      "step": 11445
    },
    {
      "entropy": 5.380377531051636,
      "epoch": 0.8908772612332231,
      "grad_norm": 1.0,
      "learning_rate": 0.0004925861202844361,
      "loss": 5.1387,
      "mean_token_accuracy": 0.18535495549440384,
      "num_tokens": 19865643.0,
      "step": 11450
    },
    {
      "entropy": 5.535952854156494,
      "epoch": 0.8912662906049407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049257906255792,
      "loss": 5.1655,
      "mean_token_accuracy": 0.18515115231275558,
      "num_tokens": 19874029.0,
      "step": 11455
    },
    {
      "entropy": 5.645742321014405,
      "epoch": 0.8916553199766583,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492572001529988,
      "loss": 5.2919,
      "mean_token_accuracy": 0.17810761630535127,
      "num_tokens": 19883347.0,
      "step": 11460
    },
    {
      "entropy": 5.544638681411743,
      "epoch": 0.8920443493483758,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925649372007469,
      "loss": 5.209,
      "mean_token_accuracy": 0.18567140847444535,
      "num_tokens": 19891822.0,
      "step": 11465
    },
    {
      "entropy": 5.475942897796631,
      "epoch": 0.8924333787200933,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925578695703045,
      "loss": 5.1903,
      "mean_token_accuracy": 0.18323710560798645,
      "num_tokens": 19899899.0,
      "step": 11470
    },
    {
      "entropy": 5.577597141265869,
      "epoch": 0.8928224080918109,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925507986387676,
      "loss": 5.3281,
      "mean_token_accuracy": 0.18126303404569627,
      "num_tokens": 19908189.0,
      "step": 11475
    },
    {
      "entropy": 5.566147327423096,
      "epoch": 0.8932114374635285,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004925437244062436,
      "loss": 5.2064,
      "mean_token_accuracy": 0.18268861174583434,
      "num_tokens": 19916166.0,
      "step": 11480
    },
    {
      "entropy": 5.528143358230591,
      "epoch": 0.8936004668352461,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004925366468728397,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18684028089046478,
      "num_tokens": 19924222.0,
      "step": 11485
    },
    {
      "entropy": 5.52174859046936,
      "epoch": 0.8939894962069637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004925295660386635,
      "loss": 5.3024,
      "mean_token_accuracy": 0.17870192229747772,
      "num_tokens": 19931925.0,
      "step": 11490
    },
    {
      "entropy": 5.583100175857544,
      "epoch": 0.8943785255786811,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004925224819038224,
      "loss": 5.2565,
      "mean_token_accuracy": 0.18239305317401885,
      "num_tokens": 19940432.0,
      "step": 11495
    },
    {
      "entropy": 5.591109132766723,
      "epoch": 0.8947675549503987,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925153944684239,
      "loss": 5.2367,
      "mean_token_accuracy": 0.18197857439517975,
      "num_tokens": 19948013.0,
      "step": 11500
    },
    {
      "entropy": 5.552106237411499,
      "epoch": 0.8951565843221163,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004925083037325754,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18869636058807374,
      "num_tokens": 19956161.0,
      "step": 11505
    },
    {
      "entropy": 5.570696830749512,
      "epoch": 0.8955456136938339,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004925012096963847,
      "loss": 5.1691,
      "mean_token_accuracy": 0.19198454171419144,
      "num_tokens": 19964533.0,
      "step": 11510
    },
    {
      "entropy": 5.597382402420044,
      "epoch": 0.8959346430655515,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004924941123599593,
      "loss": 5.2119,
      "mean_token_accuracy": 0.18112147301435472,
      "num_tokens": 19973915.0,
      "step": 11515
    },
    {
      "entropy": 5.461422252655029,
      "epoch": 0.896323672437269,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004924870117234069,
      "loss": 5.2152,
      "mean_token_accuracy": 0.18430526405572892,
      "num_tokens": 19983422.0,
      "step": 11520
    },
    {
      "entropy": 5.562084293365478,
      "epoch": 0.8967127018089865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924799077868353,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18218649327754974,
      "num_tokens": 19992185.0,
      "step": 11525
    },
    {
      "entropy": 5.545620536804199,
      "epoch": 0.8971017311807041,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004924728005503522,
      "loss": 5.0783,
      "mean_token_accuracy": 0.18648597598075867,
      "num_tokens": 20000735.0,
      "step": 11530
    },
    {
      "entropy": 5.496080160140991,
      "epoch": 0.8974907605524217,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004924656900140655,
      "loss": 5.2555,
      "mean_token_accuracy": 0.18169461488723754,
      "num_tokens": 20009758.0,
      "step": 11535
    },
    {
      "entropy": 5.5143226146697994,
      "epoch": 0.8978797899241393,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004924585761780831,
      "loss": 5.1629,
      "mean_token_accuracy": 0.17769721448421477,
      "num_tokens": 20018369.0,
      "step": 11540
    },
    {
      "entropy": 5.59825382232666,
      "epoch": 0.8982688192958569,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000492451459042513,
      "loss": 5.3005,
      "mean_token_accuracy": 0.17755273431539537,
      "num_tokens": 20027469.0,
      "step": 11545
    },
    {
      "entropy": 5.606056261062622,
      "epoch": 0.8986578486675744,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492444338607463,
      "loss": 5.2245,
      "mean_token_accuracy": 0.17962005585432053,
      "num_tokens": 20035504.0,
      "step": 11550
    },
    {
      "entropy": 5.508863162994385,
      "epoch": 0.899046878039292,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924372148730413,
      "loss": 5.1739,
      "mean_token_accuracy": 0.19141211062669755,
      "num_tokens": 20045272.0,
      "step": 11555
    },
    {
      "entropy": 5.5456148147583,
      "epoch": 0.8994359074110095,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492430087839356,
      "loss": 5.1627,
      "mean_token_accuracy": 0.1815532147884369,
      "num_tokens": 20054041.0,
      "step": 11560
    },
    {
      "entropy": 5.543204641342163,
      "epoch": 0.8998249367827271,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004924229575065152,
      "loss": 5.1177,
      "mean_token_accuracy": 0.18746805787086487,
      "num_tokens": 20062337.0,
      "step": 11565
    },
    {
      "entropy": 5.539514589309692,
      "epoch": 0.9002139661544447,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492415823874627,
      "loss": 5.2739,
      "mean_token_accuracy": 0.18309991359710692,
      "num_tokens": 20070767.0,
      "step": 11570
    },
    {
      "entropy": 5.591424417495728,
      "epoch": 0.9006029955261622,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924086869437998,
      "loss": 5.2098,
      "mean_token_accuracy": 0.18047481924295425,
      "num_tokens": 20079791.0,
      "step": 11575
    },
    {
      "entropy": 5.481707239151001,
      "epoch": 0.9009920248978798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924015467141417,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19219427406787873,
      "num_tokens": 20088458.0,
      "step": 11580
    },
    {
      "entropy": 5.493997955322266,
      "epoch": 0.9013810542695974,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923944031857613,
      "loss": 5.1994,
      "mean_token_accuracy": 0.1861426368355751,
      "num_tokens": 20097559.0,
      "step": 11585
    },
    {
      "entropy": 5.4814629554748535,
      "epoch": 0.901770083641315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004923872563587668,
      "loss": 5.0878,
      "mean_token_accuracy": 0.19254623651504515,
      "num_tokens": 20105403.0,
      "step": 11590
    },
    {
      "entropy": 5.466583919525147,
      "epoch": 0.9021591130130325,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004923801062332666,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18395482003688812,
      "num_tokens": 20115073.0,
      "step": 11595
    },
    {
      "entropy": 5.479728555679321,
      "epoch": 0.90254814238475,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004923729528093694,
      "loss": 5.2026,
      "mean_token_accuracy": 0.1841553971171379,
      "num_tokens": 20124064.0,
      "step": 11600
    },
    {
      "entropy": 5.583331441879272,
      "epoch": 0.9029371717564676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923657960871836,
      "loss": 5.2442,
      "mean_token_accuracy": 0.19219164103269576,
      "num_tokens": 20133364.0,
      "step": 11605
    },
    {
      "entropy": 5.494344425201416,
      "epoch": 0.9033262011281852,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923586360668178,
      "loss": 5.2281,
      "mean_token_accuracy": 0.1866675928235054,
      "num_tokens": 20142099.0,
      "step": 11610
    },
    {
      "entropy": 5.4702314853668215,
      "epoch": 0.9037152304999028,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923514727483807,
      "loss": 5.142,
      "mean_token_accuracy": 0.18838441967964173,
      "num_tokens": 20150421.0,
      "step": 11615
    },
    {
      "entropy": 5.574986743927002,
      "epoch": 0.9041042598716204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000492344306131981,
      "loss": 5.3034,
      "mean_token_accuracy": 0.18428612500429153,
      "num_tokens": 20159294.0,
      "step": 11620
    },
    {
      "entropy": 5.6521648406982425,
      "epoch": 0.9044932892433378,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004923371362177272,
      "loss": 5.3508,
      "mean_token_accuracy": 0.17535435110330583,
      "num_tokens": 20168082.0,
      "step": 11625
    },
    {
      "entropy": 5.544291162490845,
      "epoch": 0.9048823186150554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923299630057284,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18422658890485763,
      "num_tokens": 20176933.0,
      "step": 11630
    },
    {
      "entropy": 5.5124529838562015,
      "epoch": 0.905271347986773,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004923227864960934,
      "loss": 5.2498,
      "mean_token_accuracy": 0.18039043694734574,
      "num_tokens": 20185507.0,
      "step": 11635
    },
    {
      "entropy": 5.534951019287109,
      "epoch": 0.9056603773584906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492315606688931,
      "loss": 5.2419,
      "mean_token_accuracy": 0.18117894232273102,
      "num_tokens": 20195206.0,
      "step": 11640
    },
    {
      "entropy": 5.536805629730225,
      "epoch": 0.9060494067302082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923084235843501,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1895822137594223,
      "num_tokens": 20203290.0,
      "step": 11645
    },
    {
      "entropy": 5.5393036842346195,
      "epoch": 0.9064384361019256,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004923012371824598,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18187354505062103,
      "num_tokens": 20211808.0,
      "step": 11650
    },
    {
      "entropy": 5.572345542907715,
      "epoch": 0.9068274654736432,
      "grad_norm": 1.0,
      "learning_rate": 0.000492294047483369,
      "loss": 5.2667,
      "mean_token_accuracy": 0.17910204082727432,
      "num_tokens": 20219980.0,
      "step": 11655
    },
    {
      "entropy": 5.5504203796386715,
      "epoch": 0.9072164948453608,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004922868544871869,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1864182770252228,
      "num_tokens": 20227880.0,
      "step": 11660
    },
    {
      "entropy": 5.497947359085083,
      "epoch": 0.9076055242170784,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004922796581940227,
      "loss": 5.229,
      "mean_token_accuracy": 0.17243314981460572,
      "num_tokens": 20237023.0,
      "step": 11665
    },
    {
      "entropy": 5.611249113082886,
      "epoch": 0.907994553588796,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004922724586039855,
      "loss": 5.2695,
      "mean_token_accuracy": 0.18479921221733092,
      "num_tokens": 20246046.0,
      "step": 11670
    },
    {
      "entropy": 5.616004896163941,
      "epoch": 0.9083835829605135,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004922652557171846,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1874903380870819,
      "num_tokens": 20254392.0,
      "step": 11675
    },
    {
      "entropy": 5.444818639755249,
      "epoch": 0.9087726123322311,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004922580495337292,
      "loss": 5.1491,
      "mean_token_accuracy": 0.19083634465932847,
      "num_tokens": 20263529.0,
      "step": 11680
    },
    {
      "entropy": 5.488363361358642,
      "epoch": 0.9091616417039486,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922508400537288,
      "loss": 5.2219,
      "mean_token_accuracy": 0.18348418176174164,
      "num_tokens": 20272226.0,
      "step": 11685
    },
    {
      "entropy": 5.542445135116577,
      "epoch": 0.9095506710756662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004922436272772926,
      "loss": 5.3149,
      "mean_token_accuracy": 0.17373220771551132,
      "num_tokens": 20281084.0,
      "step": 11690
    },
    {
      "entropy": 5.553008699417115,
      "epoch": 0.9099397004473838,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922364112045301,
      "loss": 5.3019,
      "mean_token_accuracy": 0.18025858253240584,
      "num_tokens": 20290137.0,
      "step": 11695
    },
    {
      "entropy": 5.545133972167969,
      "epoch": 0.9103287298191013,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000492229191835551,
      "loss": 5.1282,
      "mean_token_accuracy": 0.19323284775018693,
      "num_tokens": 20298305.0,
      "step": 11700
    },
    {
      "entropy": 5.530434894561767,
      "epoch": 0.9107177591908189,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004922219691704645,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18429169952869415,
      "num_tokens": 20306289.0,
      "step": 11705
    },
    {
      "entropy": 5.603179264068603,
      "epoch": 0.9111067885625365,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004922147432093805,
      "loss": 5.3343,
      "mean_token_accuracy": 0.1802520915865898,
      "num_tokens": 20314826.0,
      "step": 11710
    },
    {
      "entropy": 5.663875675201416,
      "epoch": 0.911495817934254,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922075139524083,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1761022612452507,
      "num_tokens": 20323244.0,
      "step": 11715
    },
    {
      "entropy": 5.552782726287842,
      "epoch": 0.9118848473059716,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492200281399658,
      "loss": 5.2543,
      "mean_token_accuracy": 0.17747358977794647,
      "num_tokens": 20331943.0,
      "step": 11720
    },
    {
      "entropy": 5.545907878875733,
      "epoch": 0.9122738766776891,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492193045551239,
      "loss": 5.2501,
      "mean_token_accuracy": 0.18026496320962906,
      "num_tokens": 20339940.0,
      "step": 11725
    },
    {
      "entropy": 5.582332372665405,
      "epoch": 0.9126629060494067,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921858064072612,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19396570026874543,
      "num_tokens": 20348556.0,
      "step": 11730
    },
    {
      "entropy": 5.5467949390411375,
      "epoch": 0.9130519354211243,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004921785639678347,
      "loss": 5.284,
      "mean_token_accuracy": 0.1855788990855217,
      "num_tokens": 20356909.0,
      "step": 11735
    },
    {
      "entropy": 5.522295665740967,
      "epoch": 0.9134409647928419,
      "grad_norm": 1.0,
      "learning_rate": 0.000492171318233069,
      "loss": 5.1802,
      "mean_token_accuracy": 0.18770511597394943,
      "num_tokens": 20365496.0,
      "step": 11740
    },
    {
      "entropy": 5.512374591827393,
      "epoch": 0.9138299941645595,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004921640692030742,
      "loss": 5.192,
      "mean_token_accuracy": 0.18125805705785752,
      "num_tokens": 20374763.0,
      "step": 11745
    },
    {
      "entropy": 5.5617804527282715,
      "epoch": 0.914219023536277,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004921568168779603,
      "loss": 5.2463,
      "mean_token_accuracy": 0.18501508980989456,
      "num_tokens": 20383861.0,
      "step": 11750
    },
    {
      "entropy": 5.54015588760376,
      "epoch": 0.9146080529079945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004921495612578374,
      "loss": 5.2102,
      "mean_token_accuracy": 0.19265497624874114,
      "num_tokens": 20392004.0,
      "step": 11755
    },
    {
      "entropy": 5.566636085510254,
      "epoch": 0.9149970822797121,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004921423023428156,
      "loss": 5.2262,
      "mean_token_accuracy": 0.17809344977140426,
      "num_tokens": 20401078.0,
      "step": 11760
    },
    {
      "entropy": 5.676298666000366,
      "epoch": 0.9153861116514297,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004921350401330049,
      "loss": 5.363,
      "mean_token_accuracy": 0.17461358308792113,
      "num_tokens": 20409347.0,
      "step": 11765
    },
    {
      "entropy": 5.572146987915039,
      "epoch": 0.9157751410231473,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921277746285155,
      "loss": 5.181,
      "mean_token_accuracy": 0.18716677725315095,
      "num_tokens": 20417862.0,
      "step": 11770
    },
    {
      "entropy": 5.502587556838989,
      "epoch": 0.9161641703948649,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004921205058294579,
      "loss": 5.1463,
      "mean_token_accuracy": 0.18311374336481095,
      "num_tokens": 20426556.0,
      "step": 11775
    },
    {
      "entropy": 5.554633855819702,
      "epoch": 0.9165531997665823,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492113233735942,
      "loss": 5.274,
      "mean_token_accuracy": 0.17725925594568254,
      "num_tokens": 20435336.0,
      "step": 11780
    },
    {
      "entropy": 5.576182079315186,
      "epoch": 0.9169422291382999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004921059583480785,
      "loss": 5.2452,
      "mean_token_accuracy": 0.18238117545843124,
      "num_tokens": 20443884.0,
      "step": 11785
    },
    {
      "entropy": 5.522510528564453,
      "epoch": 0.9173312585100175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004920986796659775,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19682417213916778,
      "num_tokens": 20451953.0,
      "step": 11790
    },
    {
      "entropy": 5.54691219329834,
      "epoch": 0.9177202878817351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004920913976897498,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1833428680896759,
      "num_tokens": 20461480.0,
      "step": 11795
    },
    {
      "entropy": 5.584488487243652,
      "epoch": 0.9181093172534527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004920841124195055,
      "loss": 5.2438,
      "mean_token_accuracy": 0.1795766457915306,
      "num_tokens": 20470324.0,
      "step": 11800
    },
    {
      "entropy": 5.589950799942017,
      "epoch": 0.9184983466251702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004920768238553554,
      "loss": 5.2297,
      "mean_token_accuracy": 0.1787121295928955,
      "num_tokens": 20478795.0,
      "step": 11805
    },
    {
      "entropy": 5.560394668579102,
      "epoch": 0.9188873759968877,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004920695319974099,
      "loss": 5.1903,
      "mean_token_accuracy": 0.18176875114440919,
      "num_tokens": 20487922.0,
      "step": 11810
    },
    {
      "entropy": 5.5067990779876705,
      "epoch": 0.9192764053686053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004920622368457798,
      "loss": 5.1975,
      "mean_token_accuracy": 0.1936303734779358,
      "num_tokens": 20495883.0,
      "step": 11815
    },
    {
      "entropy": 5.522235488891601,
      "epoch": 0.9196654347403229,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004920549384005759,
      "loss": 5.2036,
      "mean_token_accuracy": 0.18583762496709824,
      "num_tokens": 20504403.0,
      "step": 11820
    },
    {
      "entropy": 5.495119857788086,
      "epoch": 0.9200544641120405,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004920476366619086,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19050845205783845,
      "num_tokens": 20512603.0,
      "step": 11825
    },
    {
      "entropy": 5.473958349227905,
      "epoch": 0.920443493483758,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492040331629889,
      "loss": 5.1275,
      "mean_token_accuracy": 0.18999258577823638,
      "num_tokens": 20520764.0,
      "step": 11830
    },
    {
      "entropy": 5.463704442977905,
      "epoch": 0.9208325228554756,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004920330233046277,
      "loss": 5.0985,
      "mean_token_accuracy": 0.1880857989192009,
      "num_tokens": 20529648.0,
      "step": 11835
    },
    {
      "entropy": 5.51321325302124,
      "epoch": 0.9212215522271932,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004920257116862359,
      "loss": 5.163,
      "mean_token_accuracy": 0.1889480784535408,
      "num_tokens": 20538489.0,
      "step": 11840
    },
    {
      "entropy": 5.546636533737183,
      "epoch": 0.9216105815989107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004920183967748241,
      "loss": 5.1991,
      "mean_token_accuracy": 0.1878497987985611,
      "num_tokens": 20548042.0,
      "step": 11845
    },
    {
      "entropy": 5.436558675765991,
      "epoch": 0.9219996109706283,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004920110785705037,
      "loss": 5.1087,
      "mean_token_accuracy": 0.1985417440533638,
      "num_tokens": 20556559.0,
      "step": 11850
    },
    {
      "entropy": 5.576743412017822,
      "epoch": 0.9223886403423458,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004920037570733857,
      "loss": 5.3405,
      "mean_token_accuracy": 0.1739198684692383,
      "num_tokens": 20566162.0,
      "step": 11855
    },
    {
      "entropy": 5.5404211521148685,
      "epoch": 0.9227776697140634,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004919964322835809,
      "loss": 5.0518,
      "mean_token_accuracy": 0.1928445130586624,
      "num_tokens": 20574575.0,
      "step": 11860
    },
    {
      "entropy": 5.4385003566741945,
      "epoch": 0.923166699085781,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004919891042012006,
      "loss": 5.1272,
      "mean_token_accuracy": 0.18770445883274078,
      "num_tokens": 20583394.0,
      "step": 11865
    },
    {
      "entropy": 5.483497285842896,
      "epoch": 0.9235557284574986,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004919817728263562,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18159369379281998,
      "num_tokens": 20592413.0,
      "step": 11870
    },
    {
      "entropy": 5.569636392593384,
      "epoch": 0.9239447578292161,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004919744381591586,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1906165897846222,
      "num_tokens": 20601403.0,
      "step": 11875
    },
    {
      "entropy": 5.538726663589477,
      "epoch": 0.9243337872009336,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919671001997193,
      "loss": 5.1762,
      "mean_token_accuracy": 0.18856113404035568,
      "num_tokens": 20610342.0,
      "step": 11880
    },
    {
      "entropy": 5.490780687332153,
      "epoch": 0.9247228165726512,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004919597589481497,
      "loss": 5.2173,
      "mean_token_accuracy": 0.18497272133827208,
      "num_tokens": 20619120.0,
      "step": 11885
    },
    {
      "entropy": 5.466904354095459,
      "epoch": 0.9251118459443688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491952414404561,
      "loss": 5.1606,
      "mean_token_accuracy": 0.18369911164045333,
      "num_tokens": 20628713.0,
      "step": 11890
    },
    {
      "entropy": 5.580748081207275,
      "epoch": 0.9255008753160864,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004919450665690646,
      "loss": 5.1597,
      "mean_token_accuracy": 0.18222711682319642,
      "num_tokens": 20636338.0,
      "step": 11895
    },
    {
      "entropy": 5.579671859741211,
      "epoch": 0.925889904687804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919377154417724,
      "loss": 5.1742,
      "mean_token_accuracy": 0.18615632355213166,
      "num_tokens": 20644186.0,
      "step": 11900
    },
    {
      "entropy": 5.51911997795105,
      "epoch": 0.9262789340595214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004919303610227954,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18841380923986434,
      "num_tokens": 20653672.0,
      "step": 11905
    },
    {
      "entropy": 5.488134717941284,
      "epoch": 0.926667963431239,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919230033122457,
      "loss": 5.1478,
      "mean_token_accuracy": 0.1854111894965172,
      "num_tokens": 20662936.0,
      "step": 11910
    },
    {
      "entropy": 5.518908405303955,
      "epoch": 0.9270569928029566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004919156423102345,
      "loss": 5.2568,
      "mean_token_accuracy": 0.18243345767259597,
      "num_tokens": 20671265.0,
      "step": 11915
    },
    {
      "entropy": 5.613058280944824,
      "epoch": 0.9274460221746742,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004919082780168736,
      "loss": 5.2967,
      "mean_token_accuracy": 0.17161622494459153,
      "num_tokens": 20680049.0,
      "step": 11920
    },
    {
      "entropy": 5.517641925811768,
      "epoch": 0.9278350515463918,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919009104322751,
      "loss": 5.1451,
      "mean_token_accuracy": 0.1864025816321373,
      "num_tokens": 20688411.0,
      "step": 11925
    },
    {
      "entropy": 5.548143529891968,
      "epoch": 0.9282240809181093,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004918935395565503,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1809676021337509,
      "num_tokens": 20696012.0,
      "step": 11930
    },
    {
      "entropy": 5.503843879699707,
      "epoch": 0.9286131102898268,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918861653898113,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18054557740688323,
      "num_tokens": 20704135.0,
      "step": 11935
    },
    {
      "entropy": 5.63679838180542,
      "epoch": 0.9290021396615444,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004918787879321701,
      "loss": 5.2541,
      "mean_token_accuracy": 0.17987410575151444,
      "num_tokens": 20712529.0,
      "step": 11940
    },
    {
      "entropy": 5.462137937545776,
      "epoch": 0.929391169033262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918714071837384,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19069624841213226,
      "num_tokens": 20721079.0,
      "step": 11945
    },
    {
      "entropy": 5.549027156829834,
      "epoch": 0.9297801984049796,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004918640231446282,
      "loss": 5.2716,
      "mean_token_accuracy": 0.1769583523273468,
      "num_tokens": 20730079.0,
      "step": 11950
    },
    {
      "entropy": 5.618971157073974,
      "epoch": 0.9301692277766972,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004918566358149517,
      "loss": 5.3079,
      "mean_token_accuracy": 0.18117557018995284,
      "num_tokens": 20738200.0,
      "step": 11955
    },
    {
      "entropy": 5.527298355102539,
      "epoch": 0.9305582571484147,
      "grad_norm": 1.03125,
      "learning_rate": 0.000491849245194821,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18633153140544892,
      "num_tokens": 20746663.0,
      "step": 11960
    },
    {
      "entropy": 5.5764788627624515,
      "epoch": 0.9309472865201323,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004918418512843479,
      "loss": 5.3183,
      "mean_token_accuracy": 0.17480258643627167,
      "num_tokens": 20755990.0,
      "step": 11965
    },
    {
      "entropy": 5.601742506027222,
      "epoch": 0.9313363158918498,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918344540836451,
      "loss": 5.2167,
      "mean_token_accuracy": 0.1803658574819565,
      "num_tokens": 20764985.0,
      "step": 11970
    },
    {
      "entropy": 5.541550254821777,
      "epoch": 0.9317253452635674,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918270535928244,
      "loss": 5.2209,
      "mean_token_accuracy": 0.18549701571464539,
      "num_tokens": 20773415.0,
      "step": 11975
    },
    {
      "entropy": 5.542168283462525,
      "epoch": 0.932114374635285,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004918196498119984,
      "loss": 5.2735,
      "mean_token_accuracy": 0.1800215870141983,
      "num_tokens": 20782611.0,
      "step": 11980
    },
    {
      "entropy": 5.65689058303833,
      "epoch": 0.9325034040070025,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004918122427412793,
      "loss": 5.3108,
      "mean_token_accuracy": 0.17959070801734925,
      "num_tokens": 20790859.0,
      "step": 11985
    },
    {
      "entropy": 5.610575008392334,
      "epoch": 0.9328924333787201,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004918048323807795,
      "loss": 5.2126,
      "mean_token_accuracy": 0.18499375879764557,
      "num_tokens": 20798994.0,
      "step": 11990
    },
    {
      "entropy": 5.533700275421142,
      "epoch": 0.9332814627504377,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004917974187306114,
      "loss": 5.1613,
      "mean_token_accuracy": 0.18504722267389298,
      "num_tokens": 20807343.0,
      "step": 11995
    },
    {
      "entropy": 5.557862138748169,
      "epoch": 0.9336704921221552,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004917900017908875,
      "loss": 5.1642,
      "mean_token_accuracy": 0.1907860204577446,
      "num_tokens": 20815703.0,
      "step": 12000
    },
    {
      "epoch": 0.9336704921221552,
      "eval_entropy": 5.344935700432169,
      "eval_loss": 5.24092960357666,
      "eval_mean_token_accuracy": 0.19120996321217157,
      "eval_num_tokens": 20815703.0,
      "eval_runtime": 21.784,
      "eval_samples_per_second": 1907.735,
      "eval_steps_per_second": 238.478,
      "step": 12000
    }
  ],
  "logging_steps": 5,
  "max_steps": 128520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4193949829201920.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}